- Email: [email protected]
LOCA analysis for Korean helium cooled solid breeder TBM
Fusion Engineering and Design 84 (2009) 380–384
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
LOCA analysis for Korean helium cooled solid breeder TBM Mu-Young Ahn a,∗ , Seungyon Cho a , Duck Young Ku a , Hyung-Seok Kim b , Jae-Seung Suh b a b
National Fusion Research Institute, 52, Eoeun-dong, Yuseong-gu, Daejeon 305-333, Republic of Korea ENESYS, 328, Guam-dong, Yuseong-gu, Daejeon 305-800, Republic of Korea
a r t i c l e
i n f o
Article history: Available online 26 March 2009 Keywords: TBM Accident analysis Safety analysis LOCA
a b s t r a c t One of the major ITER goals is test blanket module (TBM) program which is for the demonstration of the breeding capability that would lead to tritium self-sufficiency in a reactor and the extraction of high-grade heat suitable for electricity generation under the ITER fusion environment. While the engineering design of Korean helium cooled solid breeder (HCSB) TBM and its ancillary systems has been performed, a safety assessment on different possible accident scenarios should be carried out for the purpose of licensing. In this paper, accident analyses for several loss of coolant accident (LOCA) cases were performed in order to assess safety aspects of the TBM design using RELAP5/MOD3.2. Since the TBM forms a loop with helium cooling system (HCS) which is one of ancillary systems required for removing heat deposited in the TBM by neutron wall loading and surface heat flux from plasma, it is necessary to model the complete loop for accident analysis. In this study, the helium passage including the TBM and HCS was nodalized for each accident scenario. The TBM and HCS components were modeled as the associated heat structures provided by RELAP5 to include heat transfer across solid boundaries. Based on computational results it was found that current design of the TBM is robust from the safety point of view. © 2009 Elsevier B.V. All rights reserved.
1. Introduction An ITER test blanket module (TBM) which will correspond to a DEMO blanket is aimed at verifying the possibility of tritium selfsufficiency and the extraction of high heat using tritium breeding concepts as part of the ITER missions. The Korean helium cooled solid breeder (HCSB) TBM, one of the two concepts that Korea has proposed, utilizes a Li4 SiO4 pebble bed as breeder, a beryllium pebble bed as multiplier, a graphite pebble bed as reflector and reduced activation ferritic martensitic (RAFM) steel as structural material. As plasma facing component, beryllium armor is applied in order to protect the TBM from plasma heat flux. Breeding zone layers are optimized to provide a high tritium breeding ratio (TBR) and power production while maintaining the temperature within the design requirements by high-pressure helium coolant of 8 MPa. One of the unique features of the Korean HCSB TBM is that the amount of beryllium is considerably reduced by replacing some of them with graphite [1]. The Korean HCSB TBM is depicted in Fig. 1. While the engineering design of the Korean HCSB TBM and its ancillary systems has been performed, a safety assessment on different possible accident scenarios should be carried out for the purpose of licensing. Since the TBM employs high pressure helium to cool down the first wall (FW), side wall (SW) and breeding zone
∗ Corresponding author. Tel.: +82 42 719 1231; fax: +82 42 719 1209. E-mail address: [email protected] (M.-Y. Ahn). 0920-3796/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2009.02.047
(BZ), safety consideration is a part of the design process. Although failure mode and effects analysis (FMEA) on the TBM has not been completed, in-vessel, in-box and ex-vessel loss of coolant accident (LOCA) are considered as enveloping cases of postulated initial event (PIE) in general [2]. In the previous study, the authors reported that the TBM design is robust on the three selected LOCA cases with modeling only the TBM itself [3]. However, it is necessary to model a helium loop including both TBM and Helium Cooling System (HCS) since the TBM forms the loop with the HCS. The HCS is the main cooling system for removing heat deposited in the TBM to the secondary cooling system, which is part of the ITER Tokamak cooling water system (TCWS). The HCS consists of a recuperator, a cooler, a heater, a circulator, filters, control valves, pipelines and a pressure control unit, as shown in Fig. 2. In this paper, the accident analysis on in-vessel, in-box and ex-vessel LOCA was carried out with modeling the helium loop including the TBM and HCS. To simulate transient thermo-hydraulic behavior of the TBM for the selected scenarios, RELAP5/MOD3.2 code was used. The helium passage in the TBM and HCS was nodalized and their components were modeled as the associated heat structures provided by RELAP5 to include the heat transfer across solid boundaries. To verify the nodalization and heat structures, the TBM and HCS modeled with RELAP5 was computed assuming steady state conditions and the result in the TBM module was compared with the three-dimensional CFD result. Then the accident analyses for the three selected LOCA were performed to investigate the design robustness from the safety perspective.
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
381
Fig. 1. HCSB TBM 3D layout [3].
2. Modeling
structure models the heat transfer between the other half of the coolant in the FW and the cooling line in the BZ. To deal with heat transfer area equivalent to one-dimensional modeling of RELAP, a fouling factor was applied [4].
2.1. Nodalization Fig. 3 shows the nodalization of the TBM and HCS. For the TBM nodalization, 47 volumes and 48 junctions are used. To simulate the possible non-uniform distribution in the back manifolds, a total of 100 cooling channels in the FW and 60 tubes in the BZ are divided into 40/60 channels and 20/40 tubes respectively, and they are grouped into four pipe-components. For the HCS nodalization, 98 volumes and 98 junctions are used. In Table 1, the RELAP hydrodynamic components used for the accident analysis are explained. 2.2. Heat structure Considering the complex cooling scheme for the TBM, special treatment should be applied to the heat structure since heat flow paths are modeled in a one-dimensional manner except reflood model for using RELAP. The FW with the first breeding layer was divided and modeled into two heat structures, as in the previous work [3]. The first heat structure models the heat transfer between the plasma and a half of the coolant in the FW. The second heat
2.3. Radiation model In LOCA cases, a lack of coolant can cause a temperature increase from decay heat even in plasma shutdown thus no neutron heating. To investigate the capability of decay heat removal in LOCA cases, radiative heat transfer should be included in the analysis. In this paper a simple model using six enclosure surfaces was used for radiative heat transfer calculation as in the previous work [3]. One enclosure surface models the vacuum vessel (VV) while the others represent the common frame.
Table 1 Description of the RELAP hydrodynamic components. Accident
Component name
Component no.
Function
In-vessel LOCA
Trip valve
410
FW channel break (time trip) VV
In-box LOCA
Single volume
401
Trip valve
420 424
410
Ex-vessel LOCA
Single junction
422
Single volume
401 421 423
Trip valve
431 410
Single volume Fig. 2. HCS flow diagram [1].
430 401
SW tube break (time trip) Disk rupture in purge gas pipe (pressure trip) FW channel break (temperature trip) Connecting purge gas and purge gas pipe VV BZ + purge gas Purge gas pipe Inlet pipe break (time trip) FW channel break (temperature trip) TCWS vault VV
382
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
Fig. 3. RELAP nodalization.
3. Results of steady state analysis Calculation was performed under a steady state assumption to verify the RELAP modeling and the result was compared to the three- dimensional CFD results done by using CFX-10 [1] and the result from the previous study [3]. Nuclear power generation from the neutron wall loading of 0.78 MW/m2 and an average surface heat flux of 0.3 MW/m2 was distributed to the individual RELAP components. Detailed radial build-up and nuclear power distribution obtained from neutronics analysis can be found in [5]. Fig. 4 shows the temperature distribution in the TBM along the radial direction. It shows good agreement between the previous and the Table 2 Comparison of the computational results. Component
RELAP-present
RELAP-previous
CFD
FW Inlet/outlet temperature Inlet/outlet pressure drop
295 ◦ C/375 ◦ C 10.4 kPa
303 ◦ C/384 ◦ C 6.9 kPa
300 ◦ C/394 ◦ C 12.1 kPa
BZ and SW Inlet/outlet temperature Inlet/outlet pressure drop
376 ◦ C/478 ◦ C 536 kPa
386 ◦ C/497 ◦ C 515 kPa
394 ◦ C/494 ◦ C 522 kPa
present RELAP results. Also good agreement between the RELAP results and the CFD result is shown except in the graphite layer. The temperature profile difference in the graphite layer is because adiabatic boundary conditions were applied to the end of the graphite layer instead of taking the back manifolds into account in the case of CFD result. The temperature change and pressure drop of the helium coolant are given in Table 2. The steady state result was used for the initial conditions of the LOCA analyses. 4. Results of accident analysis For disruption load induced by LOCA, 1.8 MW/m2 , for 1 s, which is the same as the maximum disruption load for the ITER blanket FW, was applied [6]. The parameters for the simulation of three LOCA cases carried out in this study are described in Table 3. Relevant data were referred from ITER documents [6,7]. 4.1. In-vessel LOCA In-vessel LOCA occurs when one or more channels in the FW are damaged or ruptured. This leads to passive plasma shutdown with the disruption load. The progress of the accident is given in detail in Table 4. The pressurization of the VV is shown in Fig. 5. The maximum pressure in the VV was calculated to be 21.9 kPa, which is under the pressure limit of 200 kPa given by ITER [8]. Fig. 6 shows the temperature distribution along the radial direction. It is shown that the decay heat is effectively cooled only by radiative cooling. 4.2. In-box LOCA
Fig. 4. Temperature distribution in the TBM (steady state).
When the cooling tubes in the TBM box structure are broken, inbox LOCA occurs. In this case, high-pressure helium coolant flows into the BZ and then goes to tritium extraction system (TES) through the purge gas (PG) manifold. Since the TES should be maintained with low pressure, a valve system is envisaged to isolate and protect the TES from high-pressure helium coolant. In this analysis, it is assumed that a series of valves instantly protects the TES after the accident. Over-pressure in the box structure is also an issue in this case. It is foreseen to have a rupture disk which can vent the helium coolant into the VV to avoid over-pressure when pressure in the PG reaches a preset pressure. In the calculation, the preset
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
383
Table 3 Parameters for LOCA analysis. Accident
Component
Pressure
Temperature
Volume
◦
3
Break area
In-vessel LOCA
VV
1 Pa
135 C
1,090 m
3.2e−4 m2
In-box LOCA PG pipe
TBM box (BZ + PG) 100 kPa
100 kPa 450 ◦ C
450 ◦ C 1.02e−4 m3
9.22e−2 m3 2.54e−4 m2
1.15e−4 m2
Ex-vessel LOCA
TCWS vault
100 kPa
20 ◦ C
26,432 m3
6.38e−2 m2
Table 4 Accident progress of in-vessel LOCA.
Table 5 Accident progress of in-box LOCA.
Time (s)
Steps
Time (s)
Steps
t=0 0
5% cooling channels in FW break to VV 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
t=0 0 < t < = tr
5% cooling tubes break to the TBM box tr is when PG pipe pressure reaches 6 MPa; 100 % volume and surface power; decay heat; radiative cooling. 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
1
tr < t < tr + 1 tr + 1 < t
value was assumed to be 6 MPa. Details of accident progress are given in Table 5. Fig. 7 shows that plasma disruption, followed by the disk rupture, takes place at tr = 0.69 s and pressure in the box structure rapidly decreases. The temperature distribution along the radial direction is shown in Fig. 8. It shows that the decay heat removal capability is sufficient. Since this scenario could affect the ITER machine and its operation, it should be checked by the ITER organization for acceptability.
4.3. Ex-vessel LOCA Lastly, ex-vessel LOCA is considered. To simulate the most severe accident scenario, a pipe break just after the circulator in the HCS was assumed without detection. Therefore, the plasma is active until the beryllium armor melts at the melting temperature of 1283 ◦ C [6]. If the temperature of the FW reaches 1539 ◦ C which is
Fig. 5. Pressure in the VV (in-vessel LOCA).
Fig. 7. Pressure in PG pipe with rupture point (in-box LOCA).
Fig. 6. Temperature distribution (in-vessel LOCA).
Fig. 8. Temperature distribution (in-box LOCA).
384
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
Table 6 Accident progress of ex-vessel LOCA. Time (s)
Steps
t=0 0 < t < = tr
A pipe breaks in the TCWS vault tr is when the beryllium armor melts; 100% volume and surface power; decay heat; radiative cooling. 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
tr < t < tr + 1 tr + 1 < t
Fig. 11. Pressure of TCWS vault (ex-vessel LOCA).
5. Conclusions
Fig. 9. Temperature behavior on the beryllium armor (ex-vessel LOCA).
Accident analysis on the Korean HCSB TBM was performed for three LOCA cases that are considered to be enveloping cases. To simulate transient thermo-hydraulic behavior of the TBM for the selected scenarios, the helium passage including the TBM and HCS was modeled. The in-vessel LOCA leads to passive shutdown of the plasma while the pressurization of the VV is within the safety limit. The in-box LOCA also leads to passive shutdown without any safety concern with the valve system. The worst scenario is ex-vessel LOCA without detection. However, no serious damage to the TBM is found. Based on these results it was found that the TBM is designed with sufficient capability of decay heat removal for three accidental cases. Acknowledgements This work is supported by the Ministry of Education, Science and Technology (MEST) of the Republic of Korea under the ITER Project Contract References
fig. 10. temperature distribution (ex-vessel loca).
the melting temperature of Eurofer, a RAFM developed by EU-Party, then it leads to in-vessel LOCA. Details of accident progress are given in Table 6. The beryllium armor melts at tr = 93.7 s as shown in Fig. 9. The temperature of the armor reaches the maximum 1398.7 ◦ C at t = 94.7 s. Fig. 10 shows the temperature distribution along the radial direction. After reaching the maximum, the temperature on the plasma facing component slowly decreases. It shows that the decay heat is effectively cooled by radiative heat transfer. The maximum temperature of the RAFM is calculated to be 1343.2 ◦ C at t = 94.9 s. Therefore, no melting of the FW occurs and it does not proceed to in-vessel LOCA. The pressurization of the TCWS vault is negligible as shown in Fig. 11.
[1] S. Cho, M.Y. Ahn, D.H. Kim, E.S. Lee, S. Yun, N.Z. Cho, K.J. Jung, Current status of design and analysis of Korea helium cooled solid breeder test blanket module, Fusion Eng. Des. 83 (2008). [2] ITER Test Blanket Working Group, Progress Report 2006 on TBM Safety & Licensing, 2007, see http://user.iter.org/?uid=2FT4Q8. [3] M.Y. Ahn, S. Cho, D.H. Kim, E.S. Lee, H.S. Kim, J.S. Suh, S. Yun, N.Z. Cho, Preliminary safety analysis of korea helium cooled solid breeder test blanket module, Fus. Eng. Des. 83 (2008). [4] RELAP5 Code Development Team, RELAP5/MOD3 Code Manual, 1995. [5] S. Yun, N.Z. Cho, M.Y. Ahn, S. Cho, Depletion analysis of a solid type blanket design for ITER, Fusion Eng. Des. 83 (2008). [6] ITER, Safety Analysis Data List, Version 5.1.1, 2006, see http://user.iter. org/?uid=24LSAE. [7] ITER, Project Integration Document, Version 3.0, 2007, see http://user. iter.org/?uid=2234RH. [8] ITER, Design Description Document of WBS 1.5 Vacuum Vessel, G 15 DDD 4 R0.2, see http://user.iter.org/?uid=22FPWQ.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
LOCA analysis for Korean helium cooled solid breeder TBM Mu-Young Ahn a,∗ , Seungyon Cho a , Duck Young Ku a , Hyung-Seok Kim b , Jae-Seung Suh b a b
National Fusion Research Institute, 52, Eoeun-dong, Yuseong-gu, Daejeon 305-333, Republic of Korea ENESYS, 328, Guam-dong, Yuseong-gu, Daejeon 305-800, Republic of Korea
a r t i c l e
i n f o
Article history: Available online 26 March 2009 Keywords: TBM Accident analysis Safety analysis LOCA
a b s t r a c t One of the major ITER goals is test blanket module (TBM) program which is for the demonstration of the breeding capability that would lead to tritium self-sufficiency in a reactor and the extraction of high-grade heat suitable for electricity generation under the ITER fusion environment. While the engineering design of Korean helium cooled solid breeder (HCSB) TBM and its ancillary systems has been performed, a safety assessment on different possible accident scenarios should be carried out for the purpose of licensing. In this paper, accident analyses for several loss of coolant accident (LOCA) cases were performed in order to assess safety aspects of the TBM design using RELAP5/MOD3.2. Since the TBM forms a loop with helium cooling system (HCS) which is one of ancillary systems required for removing heat deposited in the TBM by neutron wall loading and surface heat flux from plasma, it is necessary to model the complete loop for accident analysis. In this study, the helium passage including the TBM and HCS was nodalized for each accident scenario. The TBM and HCS components were modeled as the associated heat structures provided by RELAP5 to include heat transfer across solid boundaries. Based on computational results it was found that current design of the TBM is robust from the safety point of view. © 2009 Elsevier B.V. All rights reserved.
1. Introduction An ITER test blanket module (TBM) which will correspond to a DEMO blanket is aimed at verifying the possibility of tritium selfsufficiency and the extraction of high heat using tritium breeding concepts as part of the ITER missions. The Korean helium cooled solid breeder (HCSB) TBM, one of the two concepts that Korea has proposed, utilizes a Li4 SiO4 pebble bed as breeder, a beryllium pebble bed as multiplier, a graphite pebble bed as reflector and reduced activation ferritic martensitic (RAFM) steel as structural material. As plasma facing component, beryllium armor is applied in order to protect the TBM from plasma heat flux. Breeding zone layers are optimized to provide a high tritium breeding ratio (TBR) and power production while maintaining the temperature within the design requirements by high-pressure helium coolant of 8 MPa. One of the unique features of the Korean HCSB TBM is that the amount of beryllium is considerably reduced by replacing some of them with graphite [1]. The Korean HCSB TBM is depicted in Fig. 1. While the engineering design of the Korean HCSB TBM and its ancillary systems has been performed, a safety assessment on different possible accident scenarios should be carried out for the purpose of licensing. Since the TBM employs high pressure helium to cool down the first wall (FW), side wall (SW) and breeding zone
∗ Corresponding author. Tel.: +82 42 719 1231; fax: +82 42 719 1209. E-mail address: [email protected] (M.-Y. Ahn). 0920-3796/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2009.02.047
(BZ), safety consideration is a part of the design process. Although failure mode and effects analysis (FMEA) on the TBM has not been completed, in-vessel, in-box and ex-vessel loss of coolant accident (LOCA) are considered as enveloping cases of postulated initial event (PIE) in general [2]. In the previous study, the authors reported that the TBM design is robust on the three selected LOCA cases with modeling only the TBM itself [3]. However, it is necessary to model a helium loop including both TBM and Helium Cooling System (HCS) since the TBM forms the loop with the HCS. The HCS is the main cooling system for removing heat deposited in the TBM to the secondary cooling system, which is part of the ITER Tokamak cooling water system (TCWS). The HCS consists of a recuperator, a cooler, a heater, a circulator, filters, control valves, pipelines and a pressure control unit, as shown in Fig. 2. In this paper, the accident analysis on in-vessel, in-box and ex-vessel LOCA was carried out with modeling the helium loop including the TBM and HCS. To simulate transient thermo-hydraulic behavior of the TBM for the selected scenarios, RELAP5/MOD3.2 code was used. The helium passage in the TBM and HCS was nodalized and their components were modeled as the associated heat structures provided by RELAP5 to include the heat transfer across solid boundaries. To verify the nodalization and heat structures, the TBM and HCS modeled with RELAP5 was computed assuming steady state conditions and the result in the TBM module was compared with the three-dimensional CFD result. Then the accident analyses for the three selected LOCA were performed to investigate the design robustness from the safety perspective.
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
381
Fig. 1. HCSB TBM 3D layout [3].
2. Modeling
structure models the heat transfer between the other half of the coolant in the FW and the cooling line in the BZ. To deal with heat transfer area equivalent to one-dimensional modeling of RELAP, a fouling factor was applied [4].
2.1. Nodalization Fig. 3 shows the nodalization of the TBM and HCS. For the TBM nodalization, 47 volumes and 48 junctions are used. To simulate the possible non-uniform distribution in the back manifolds, a total of 100 cooling channels in the FW and 60 tubes in the BZ are divided into 40/60 channels and 20/40 tubes respectively, and they are grouped into four pipe-components. For the HCS nodalization, 98 volumes and 98 junctions are used. In Table 1, the RELAP hydrodynamic components used for the accident analysis are explained. 2.2. Heat structure Considering the complex cooling scheme for the TBM, special treatment should be applied to the heat structure since heat flow paths are modeled in a one-dimensional manner except reflood model for using RELAP. The FW with the first breeding layer was divided and modeled into two heat structures, as in the previous work [3]. The first heat structure models the heat transfer between the plasma and a half of the coolant in the FW. The second heat
2.3. Radiation model In LOCA cases, a lack of coolant can cause a temperature increase from decay heat even in plasma shutdown thus no neutron heating. To investigate the capability of decay heat removal in LOCA cases, radiative heat transfer should be included in the analysis. In this paper a simple model using six enclosure surfaces was used for radiative heat transfer calculation as in the previous work [3]. One enclosure surface models the vacuum vessel (VV) while the others represent the common frame.
Table 1 Description of the RELAP hydrodynamic components. Accident
Component name
Component no.
Function
In-vessel LOCA
Trip valve
410
FW channel break (time trip) VV
In-box LOCA
Single volume
401
Trip valve
420 424
410
Ex-vessel LOCA
Single junction
422
Single volume
401 421 423
Trip valve
431 410
Single volume Fig. 2. HCS flow diagram [1].
430 401
SW tube break (time trip) Disk rupture in purge gas pipe (pressure trip) FW channel break (temperature trip) Connecting purge gas and purge gas pipe VV BZ + purge gas Purge gas pipe Inlet pipe break (time trip) FW channel break (temperature trip) TCWS vault VV
382
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
Fig. 3. RELAP nodalization.
3. Results of steady state analysis Calculation was performed under a steady state assumption to verify the RELAP modeling and the result was compared to the three- dimensional CFD results done by using CFX-10 [1] and the result from the previous study [3]. Nuclear power generation from the neutron wall loading of 0.78 MW/m2 and an average surface heat flux of 0.3 MW/m2 was distributed to the individual RELAP components. Detailed radial build-up and nuclear power distribution obtained from neutronics analysis can be found in [5]. Fig. 4 shows the temperature distribution in the TBM along the radial direction. It shows good agreement between the previous and the Table 2 Comparison of the computational results. Component
RELAP-present
RELAP-previous
CFD
FW Inlet/outlet temperature Inlet/outlet pressure drop
295 ◦ C/375 ◦ C 10.4 kPa
303 ◦ C/384 ◦ C 6.9 kPa
300 ◦ C/394 ◦ C 12.1 kPa
BZ and SW Inlet/outlet temperature Inlet/outlet pressure drop
376 ◦ C/478 ◦ C 536 kPa
386 ◦ C/497 ◦ C 515 kPa
394 ◦ C/494 ◦ C 522 kPa
present RELAP results. Also good agreement between the RELAP results and the CFD result is shown except in the graphite layer. The temperature profile difference in the graphite layer is because adiabatic boundary conditions were applied to the end of the graphite layer instead of taking the back manifolds into account in the case of CFD result. The temperature change and pressure drop of the helium coolant are given in Table 2. The steady state result was used for the initial conditions of the LOCA analyses. 4. Results of accident analysis For disruption load induced by LOCA, 1.8 MW/m2 , for 1 s, which is the same as the maximum disruption load for the ITER blanket FW, was applied [6]. The parameters for the simulation of three LOCA cases carried out in this study are described in Table 3. Relevant data were referred from ITER documents [6,7]. 4.1. In-vessel LOCA In-vessel LOCA occurs when one or more channels in the FW are damaged or ruptured. This leads to passive plasma shutdown with the disruption load. The progress of the accident is given in detail in Table 4. The pressurization of the VV is shown in Fig. 5. The maximum pressure in the VV was calculated to be 21.9 kPa, which is under the pressure limit of 200 kPa given by ITER [8]. Fig. 6 shows the temperature distribution along the radial direction. It is shown that the decay heat is effectively cooled only by radiative cooling. 4.2. In-box LOCA
Fig. 4. Temperature distribution in the TBM (steady state).
When the cooling tubes in the TBM box structure are broken, inbox LOCA occurs. In this case, high-pressure helium coolant flows into the BZ and then goes to tritium extraction system (TES) through the purge gas (PG) manifold. Since the TES should be maintained with low pressure, a valve system is envisaged to isolate and protect the TES from high-pressure helium coolant. In this analysis, it is assumed that a series of valves instantly protects the TES after the accident. Over-pressure in the box structure is also an issue in this case. It is foreseen to have a rupture disk which can vent the helium coolant into the VV to avoid over-pressure when pressure in the PG reaches a preset pressure. In the calculation, the preset
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
383
Table 3 Parameters for LOCA analysis. Accident
Component
Pressure
Temperature
Volume
◦
3
Break area
In-vessel LOCA
VV
1 Pa
135 C
1,090 m
3.2e−4 m2
In-box LOCA PG pipe
TBM box (BZ + PG) 100 kPa
100 kPa 450 ◦ C
450 ◦ C 1.02e−4 m3
9.22e−2 m3 2.54e−4 m2
1.15e−4 m2
Ex-vessel LOCA
TCWS vault
100 kPa
20 ◦ C
26,432 m3
6.38e−2 m2
Table 4 Accident progress of in-vessel LOCA.
Table 5 Accident progress of in-box LOCA.
Time (s)
Steps
Time (s)
Steps
t=0 0
5% cooling channels in FW break to VV 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
t=0 0 < t < = tr
5% cooling tubes break to the TBM box tr is when PG pipe pressure reaches 6 MPa; 100 % volume and surface power; decay heat; radiative cooling. 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
1
tr < t < tr + 1 tr + 1 < t
value was assumed to be 6 MPa. Details of accident progress are given in Table 5. Fig. 7 shows that plasma disruption, followed by the disk rupture, takes place at tr = 0.69 s and pressure in the box structure rapidly decreases. The temperature distribution along the radial direction is shown in Fig. 8. It shows that the decay heat removal capability is sufficient. Since this scenario could affect the ITER machine and its operation, it should be checked by the ITER organization for acceptability.
4.3. Ex-vessel LOCA Lastly, ex-vessel LOCA is considered. To simulate the most severe accident scenario, a pipe break just after the circulator in the HCS was assumed without detection. Therefore, the plasma is active until the beryllium armor melts at the melting temperature of 1283 ◦ C [6]. If the temperature of the FW reaches 1539 ◦ C which is
Fig. 5. Pressure in the VV (in-vessel LOCA).
Fig. 7. Pressure in PG pipe with rupture point (in-box LOCA).
Fig. 6. Temperature distribution (in-vessel LOCA).
Fig. 8. Temperature distribution (in-box LOCA).
384
M.-Y. Ahn et al. / Fusion Engineering and Design 84 (2009) 380–384
Table 6 Accident progress of ex-vessel LOCA. Time (s)
Steps
t=0 0 < t < = tr
A pipe breaks in the TCWS vault tr is when the beryllium armor melts; 100% volume and surface power; decay heat; radiative cooling. 100% volume and surface power; additional disruption load 1.8 MW/m2 ; decay heat; radiative cooling. Plasma quench; decay heat; radiative cooling.
tr < t < tr + 1 tr + 1 < t
Fig. 11. Pressure of TCWS vault (ex-vessel LOCA).
5. Conclusions
Fig. 9. Temperature behavior on the beryllium armor (ex-vessel LOCA).
Accident analysis on the Korean HCSB TBM was performed for three LOCA cases that are considered to be enveloping cases. To simulate transient thermo-hydraulic behavior of the TBM for the selected scenarios, the helium passage including the TBM and HCS was modeled. The in-vessel LOCA leads to passive shutdown of the plasma while the pressurization of the VV is within the safety limit. The in-box LOCA also leads to passive shutdown without any safety concern with the valve system. The worst scenario is ex-vessel LOCA without detection. However, no serious damage to the TBM is found. Based on these results it was found that the TBM is designed with sufficient capability of decay heat removal for three accidental cases. Acknowledgements This work is supported by the Ministry of Education, Science and Technology (MEST) of the Republic of Korea under the ITER Project Contract References
fig. 10. temperature distribution (ex-vessel loca).
the melting temperature of Eurofer, a RAFM developed by EU-Party, then it leads to in-vessel LOCA. Details of accident progress are given in Table 6. The beryllium armor melts at tr = 93.7 s as shown in Fig. 9. The temperature of the armor reaches the maximum 1398.7 ◦ C at t = 94.7 s. Fig. 10 shows the temperature distribution along the radial direction. After reaching the maximum, the temperature on the plasma facing component slowly decreases. It shows that the decay heat is effectively cooled by radiative heat transfer. The maximum temperature of the RAFM is calculated to be 1343.2 ◦ C at t = 94.9 s. Therefore, no melting of the FW occurs and it does not proceed to in-vessel LOCA. The pressurization of the TCWS vault is negligible as shown in Fig. 11.
[1] S. Cho, M.Y. Ahn, D.H. Kim, E.S. Lee, S. Yun, N.Z. Cho, K.J. Jung, Current status of design and analysis of Korea helium cooled solid breeder test blanket module, Fusion Eng. Des. 83 (2008). [2] ITER Test Blanket Working Group, Progress Report 2006 on TBM Safety & Licensing, 2007, see http://user.iter.org/?uid=2FT4Q8. [3] M.Y. Ahn, S. Cho, D.H. Kim, E.S. Lee, H.S. Kim, J.S. Suh, S. Yun, N.Z. Cho, Preliminary safety analysis of korea helium cooled solid breeder test blanket module, Fus. Eng. Des. 83 (2008). [4] RELAP5 Code Development Team, RELAP5/MOD3 Code Manual, 1995. [5] S. Yun, N.Z. Cho, M.Y. Ahn, S. Cho, Depletion analysis of a solid type blanket design for ITER, Fusion Eng. Des. 83 (2008). [6] ITER, Safety Analysis Data List, Version 5.1.1, 2006, see http://user.iter. org/?uid=24LSAE. [7] ITER, Project Integration Document, Version 3.0, 2007, see http://user. iter.org/?uid=2234RH. [8] ITER, Design Description Document of WBS 1.5 Vacuum Vessel, G 15 DDD 4 R0.2, see http://user.iter.org/?uid=22FPWQ.