Ac

MS

h

••••

a

ARRA

1

ueP

t

h0

Nuclear Engineering and Design 272 (2014) 45–52

Contents lists available at ScienceDirect

Nuclear Engineering and Design

jou rn al hom ep age: www.elsev ier .com/ locate /nucengdes

ccident analyses for china pressurizer reactor with an innovativeonceptual design of passive residual heat removal system

ingjun Wang, Dalin Zhang ∗, Suizheng Qiu, Wenxi Tian, Guanghui Sutate Key Laboratory of Multiphase Flow in Power Engineering, School of Nuclear Science and Technology, Xi’an Jiaotong University, Xi’an 710049, China

i g h l i g h t s

An innovative passive system design concept was proposed.Remote trigger control was adopted in the new passive system.Transient characteristic of new PRHRS was studied during normal operation, SBO and FLB.New design type is successful.

r t i c l e i n f o

rticle history:eceived 12 July 2013eceived in revised form 15 January 2014ccepted 28 January 2014

a b s t r a c t

An innovative passive safety system design concept for CPR 1000 nuclear power plant (NPP), proposedbased on the original passive residual heat removal system (PRHRS) design, is presented and studied inthis paper. The new type PRHRS avoids use of trigger valve inside the containment and adds an externalwater tank as the trip of PRHRS operation. The new design concept can realize functions of system remotecontrol and trigger, while the external water tank can provide emergency water for steam generator (SG)in case of loss of feed water accident. Also, the advantages to use a valve and a tank located externally tocontainment are including the more reliability in implementing valve in pipes at low pressure, possibilityoperating manually the valve to start the system, refilling the tank in case of long time cooling andmaintaining the valve easily. Best-estimate transient simulation code Relap5/MOD3.4 is applied to studythe behavior of new design PRHRS and transient characteristics of primary loop system during normalcondition and accident conditions. The transient processes of Station Black-Out (SBO) and Feed-waterLine Break (FLB) accidents are studied to verify the function of new design PRHRS. Results show that theeffect of valve position change from inside to outside containment can be ignored for the new designPRHRS in normal operation and the external valve can trip the operation of PRHRS in case of accidents.

The new design PRHRS can remove core residual heat from primary loop effectively through establishingthe stable natural circulations in the primary loop and PRHRS loop when the SBO and FLB accidents occur.It also can supply the emergency water to the SG in the event of FLB accident and realize safe reactorshutdown. Results indicate that the new design concept PRHRS in this work is potential and successfulin future application.

. Introduction

The innovative safety systems often rely on passive safety meas-res like gravity injection or natural circulation and do not needxternal input energy to operate, improving inherent safety for

ressurized Water Reactor (PWR).

For China Pressurized Reactor (CPR1000), much research relatedo passive safety system has been performed in recent years. Zhang

∗ Corresponding author.E-mail address: dlzhang@mail.xjtu.edu.cn (D. Zhang).

ttp://dx.doi.org/10.1016/j.nucengdes.2014.01.019029-5493/© 2014 Elsevier B.V. All rights reserved.

© 2014 Elsevier B.V. All rights reserved.

et al. (2011) has designed a type of secondary emergency passiveresidual heat removal system (PRHRS) for CPR1000 and studiedthe transient characteristics of the primary loop system and thePRHRS in the event of a feed line break (FLB) or loss of heat sinkaccident. Another type of secondary PRHRS has been designed andstudied in the case of Station Black-Out (SBO) accident for CPR1000by Wang et al. (2012). The results of the study demonstrated thatthe designed secondary PRHRS was effective and meaningful to

CPR1000. In addition, experimental research on passive safety sys-tems was carried out for CPR1000 NPP (Xiao et al., 2003), the resultsof which indicated that the passive system was effective and reli-able in principle and could meet the design and safety requirement.

4 eering and Design 272 (2014) 45–52

iptgattgp

mwopicoe

fcgnpoPdnaPtPdoBs

2

tSoadttSHtpttfttpaTw

aCe

Table 1Main parameters of the new PRHRS.

Parameters Values

Area of ascending pipe 0.04917 m2

HX area 300 m2

Area of descending pipe 0.00836 m2

Area of HX tank 10 m2

Height of HX tank 12 mVolume of HX tank 98.88 m3

Area of external water tank 113.04 m2

Height of external water tank 12 mFlow area of valve between two tanks 0.1256 m2

valve used inside the containment. Based on the steady calculationmodel, the CPR1000 transient analysis and the function character-istic of new PRHRS are studied in the following sections.

Table 2Steady calculation results comparison with design value.

Parameters Value in thiswork

Designvalue

Error

Primary loop pressure (MPa) 15.709 15.500 1.35%

6 M. Wang et al. / Nuclear Engin

For above types of designs, the trigger valve is necessary to real-ze the passive system function. The trigger valve is closed andrevents the heat transfer by insulating the heat exchanger (HX)ubes with overheated steam during normal operation. If the trig-er valve is opened, the steam is condensed in the passive safety HXnd cold water flows from the HX to the SG. In this case, an effec-ive natural circulation is established allowing passive removal ofhe decay heat. Preliminary calculations showed the use of a trig-er valve brought significant system uncertainty and constructionroblems (Meloni and Pignatel, 1998; Meloni, 1998).

To overcome this problem, a new system configuration, whichainly consists of a HX pool, external pool and a heat exchanger,as proposed by ENEA and SIET (Achilli et al., 2002). The research

n heat removal by in-pool immersed heat exchangers has beenerformed at SIET laboratories using the PERSEO facility by exper-

mental method (Ferri et al., 2005). In addition, the best estimateode Relap5 was utilized as design instrument during all phasesf this project. Results generally show a good agreement betweenxperimental and calculated data and identify some code limits.

As discussed above, the design types for CPR1000 among dif-erent works always similarly contain a trigger valve in the pipeonnected to the SG or reactor, increasing system uncertaintyreatly. The assessment of the reliability of passive systems is a sig-ificant issue to be solved for their extensive use in future nuclearower plants (Zio and Pedroni, 2009). Therefore, in this study, basedn the original type of SG secondary PRHRS, the new conceptualRHRS which avoids the use of trigger valve connected to the SG iseveloped. An external water tank is added to the PRHRS. An exter-al valve connects the external water tank and HX tank filled withir initially. Furthermore, the model of the primary loop and theRHRS were established using Relap5/MOD3.4 (2001a,b) to inves-igate the core residual heat removal capability of the new conceptRHRS. To investigate the function of PRHRS, the transient acci-ents, leading to the loss of secondary side feed water and shortf SG water volume, are selected to study. In this work, the Stationlack-Out (SBO) and Feed water Line Break (FLB) accidents wereimulated.

. New PRHRS conceptual design

The new concept PRHRS, which adds an external water tank ashe heat sink and trip of PRHRS operation, is also connected to theG secondary side and consists of three independent loops. Theriginal design schematic diagram and new PRHRS design diagramre shown in Fig. 1. Compared with the original design, the greatifference is that the valve is avoided inside the containment inhe new design, increasing the reliability of PRHRS significantly. Inhe original PRHRS type, the trigger valve connects the HX withG, controlling the start of PRHRS. The steam in the SG flows toX through ascending pipe and is condensed in the HX. At last,

he condensed coolant flows back to SG through the descendingipe. The heat is transferred to the water in the water tank, takinghe primary loop heat out. For the new type PRHRS, there are twoanks, which are HX tank and external water tank. The HX tank isull of air in the normal condition and the external valve will triphe operation of the PRHRS, in which case the HX is filled up andhe heat transfer begins. Furthermore, the external water tank canrovide the emergency water to SG in case of loss of feed waterccident. So the start valve is tripped based on the accident signals.he feed water valve is tripped when the SG needs to be suppliedater.

Each new PRHRS loop is composed of a steam generator, an HX,n HX tank and an external water tank. Based on the power ofPR1000 NPP and capacity of each new PRHRS, the main param-ters of new designed PRHRS are shown in Table 1.

HX active length 3.0 mFlow area of valve between external water tank and SG 0.005 m2

3. Relap5 modeling

The best-estimate transient simulation code Relap5/Mod3.4 isutilized to carry out the calculations presented here. Only one of thethree loops and corresponding PRHRS Relap5 node is shown dueto the system symmetry. The reactor vessel, reactor core, pressur-izer, reactor coolant pump, accumulators and steam generators aremodeled in detail. Other important auxiliary systems are modeledusing time-dependent volume (TMDPVOL) and time-dependentjunction (TMDPJUN) in this study.

As shown in Fig. 2, for the PRHRS, the HX, saturated steamascending pipe, condensed water descending pipe, HX tank, exter-nal water tank and corresponding pipes and valves are simulateddetailedly by different Relap5 models. The HX tubes heat transferis modeled by heat structure. The external isolation valve trip is theswitch for controlling PRHRS operation.

4. Results and discussions

The characteristics of the new PRHRS and CPR1000 NPP primaryloop are studied in normal operational conditions and accidentconditions in this section.

4.1. Steady analysis and PRHRS characteristic in normaloperation

The steady state analysis is studied firstly for evaluating theCPR1000 Relap5 model and the effects of new PRHRS without valve.The main parameters comparison between Relap5 code steady cal-culation results incorporating new PRHRS design and CPR1000design data under normal operational condition are shown inTable 2.It can be seen from comparison results in Table 2 that theRelap5 calculation values of important reactor thermal hydraulicparameters, i.e. primary pressure, primary loop flow rate, SGsecondary side pressure, cold leg temperature and hot leg temper-ature, are in good agreement with the design values. The errors areall in the acceptable range. Therefore, the new PRHRS has almost noeffect on the CPR1000 NPP in normal operation condition without

Reactor outlet temperature (◦C) 329.852 329.800 0.016%Reactor inlet temperature (◦C) 293.858 292.400 0.5%Thermal design flow rate (kg/s) 4661.273 4706.560 1.0%SG pressure (MPa) 6.908 6.890 0.3%

M. Wang et al. / Nuclear Engineering and Design 272 (2014) 45–52 47

f PRHR

4

lafl

Fig. 1. Schematic diagram o

.2. SBO analysis

The characteristics of primary loop and new PRHRS are ana-yzed in case of SBO accident in this section. The SBO accident isssumed to occur at 0 s. The external valve opens and the waterows from the external water tank to HX tank, which makes the

S original and new designs.

new PRHRS operation start. Another assumption is that the dieselgenerator does not work, considering the worst condition. The main

steam is tripped at the same time to protect the turbine. The mainpump begin to run out and the control rods drop caused by thegravity, leading to a 5 s delay of the reactor shutdown. Then thePRHRS receives the start signal because of reactor shutdown and

48 M. Wang et al. / Nuclear Engineering and Design 272 (2014) 45–52

the n

tStwmF

Fig. 2. Relap5 nodalization of

he start valve opens, making the PRHRS run. The accident logic ofBO is shown in Table 3.The characteristic comparisons of impor-

ant CPR1000 NPP thermal hydraulics parameters between modelsith the PRHRS and without the PRHRS are shown in Figs. 3–5. Theain parametric characteristics of the new PRHRS are shown in

igs. 6–9.

ew PRHRS and primary loop.

Fig. 3 shows the variation of primary loop pressure with time.As can be seen in the figure, the first pressure peak occurs soon

after the SBO scenario starts, caused by the main pumps runningout at the beginning of accident. Also, the transient peak value ofreactor coolant system pressure is less than 110% design pressure(16.72 MPa) and then decreases to a lower value with the new

M. Wang et al. / Nuclear Engineering and Design 272 (2014) 45–52 49

Table 3SBO accident logic.

Event Time

SBO occurs 0 sMain pump runs out 0 sFeed-water stop 0 sMain steam system trip 0 sShutdown of reactor 5 s (control rods drop down)PRHRS start valve open 5 s (shutdown signal)

Fig. 3. Variation of primary loop pressure vs. time.

Patldv

fotwh

sw

Fig. 5. Variation of hot leg temperature and saturated temperature vs. time.

Fig. 6. Variation of tank water level vs. time.

the hot leg temperature will reach the saturated temperature at the

Fig. 4. Variation of SG secondary side pressure vs. time.

RHRS operation when the SBO accident occurs. However, in thebsence of PRHRS, the primary loop pressure will decrease due tohe SG protection valve open which has a feedback on the primaryoop and the reactor shutdown at the beginning, but it will notecrease to a safety value and the pressure will increase to a highalue, which is dangerous for reactor operation.

The SG secondary side pressure will increase with the loss ofeed-water firstly, and then decrease due to the SG protection valvepen, as shown in Fig. 4. The SG secondary side pressure will con-inue to decrease because of the PRHRS operation. On the contrary,ithout the PRHRS the SG secondary side pressure will maintain aigh value due to loss of heat sink, as shown in Fig. 4.

Fig. 5 illustrates the variations of the hot leg temperature andaturated temperature of the primary loop. The hot leg temperatureill not reach the saturated temperature due to the PRHRS working

Fig. 7. Variation of flow rate from external tank to HX tank vs. time.

in the whole process and will decrease to a low and safe value, while

later stage without PRHRS.Figs. 6 and 7 illustrate the variations of flow rate from external

tank to HX tank and tank water level with time. The external valve

50 M. Wang et al. / Nuclear Engineering and Design 272 (2014) 45–52

Fig. 8. Variation of PRHRS HX heat transfer vs. time.

odaHoftftasipec

pe

4

tata

Table 4FLB accident logic.

Event Time

FLB occurs 0 sMain pump runs out SG low pressure signal (3.0 MPa)Feed-water stop 0 sMain steam system trip SG low–low pressure signal (0.15 MPa)Shutdown of reactor SG low pressure signal (3.0 MPa)PRHRS feed water valve open SG low–low pressure signal (0.15 MPa)PRHRS start valve open Reactor shutdown signal

Fig. 10. Variation of primary loop pressure vs. time.

Fig. 9. Variation of flow rate of PRHRS vs. time.

pens and the water begins to flow from external tank to HX tankue to gravity. The water level of external water tank begins to dropnd the HX water volume increases. It needs about 140 s to fill theX tank and it begins to stay in a steady value for a while. The strongscillation in Fig. 7 at the beginning is because the strong heat trans-er from HX inside to the HX tank. The strong heat transfer leads tohe water evaporation in HX tank and makes the flow rate oscillaterom external tank to HX tank. The heat transfer begins in the HXank and the natural circulation is established in the new PRHRSs well. The heat flux jumps to a high value and then decreaseslowly because of the reactor shutdown and power decline, whichs shown in Fig. 8. The flow rates of ascending pipe and descendingipe in PRHRS are shown in Fig. 9. The stable natural circulation isstablished in the new PRHRS at about 1000 s, which ensures theore decay residual heat remove from the primary loop.

In conclusion, calculation results show that the new PRHRS canrovide the final heat sink and remove the residual heat of CPR1000ffectively in case of SBO accident.

.3. FLB analysis

The CPR1000 SG pressure begins to decrease and the reac-

or will trip due to SG low pressure, when an assumed breakccident only occurs in the main feed-water line of the loop A inhe NPP, namely feed line break (FLB). The break area is assumeds 0.4162 m2 in this study. Then the reactor coolant pumps and

Fig. 11. Variation of hot leg temperature vs. time.

the main feed-water pumps run out due to reactor shutdown. Theactive auxiliary feed water is assumed to fail in most conservativeconditions. The main steam valve turns off due to the steam turbinetripping. The PRHRS start valve opens when it receives the reactorshutdown signal, while the PRHRS feed water valve opens when itreceives the SG low–low pressure signal. The detailed FLB accidentlogic is shown in Table 4. The main parametric characteristics ofprimary loop and new PRHRS are shown in Figs. 10–19.

Fig. 10 shows the variation of primary loop pressure withtime. As can be seen in the figure, the new PRHRS can realizethe reactor safety shutdown. However, the primary loop pressure

without new PRHRS operation will maintain a high value whichwill threaten the reactor safety. Fig. 11 illustrates the variations ofthe hot leg temperature and saturated temperature of the primaryloop with time. The hot leg temperature will reach the saturated

M. Wang et al. / Nuclear Engineering and Design 272 (2014) 45–52 51

Fig. 12. Variation of flow rate from external tank to HX tank vs. time.

Fig. 13. Variation of flow rate from external tank to SG vs. time.

ttSc

Fig. 15. Variation of PRHRS HX heat transfer vs. time.

Fig. 16. Variation of flow rate of loop A PRHRS vs. time.

decrease to a low and safe level.

Fig. 14. Variation of tank water level vs. time.

emperature due to the loss of SG secondary side feed water. On

he other hand, the new PRHRS can provide emergency water toG as well as establish stable natural circulation, removing theore heat after accident. In this case, the hot leg temperature will

Fig. 17. Variation of flow rate of loop B PRHRS vs. time.

not reach the saturated temperature in the whole process and will

Figs. 12 and 13 show the variations of flow rate from externalwater tank to HX tank and to SG, respectively. The flow rate willoscillate during the process because of the vaporization in HX tank

52 M. Wang et al. / Nuclear Engineering

Fig. 18. Variation of discharge flow rate vs. time.

aFeetcdv4f

wlwoi

Ttw

ltI

Fig. 19. Variation of SG water volume vs. time.

nd pressure change in SG, leading to a pressure difference change.urthermore, the flow rates results in the water levels change inxternal tank, HX tank and SG, as shown in Figs. 14 and 19. Thexternal water tank water volume decreases with an increase in HXank. The water level difference between two tanks almost remainsonstant caused by the pressure loss. The break SG water volumeecreases sharply to a low value because of the FLB and initialaporization. Then the water volume begins to increase from about500 s due to the heat transfer decline in SG and water injectionrom external tank.

Fig. 15 shows the variation of loop A PRHRS HX heat transferith time. The flow rates of ascending pipe and descending pipe in

oop A PRHRS are shown in Fig. 16. The FLB accident leads to the SGater loss and weakens the heat transfer. Fig. 17 shows variation

f flow rate of loop B PRHRS with time. Obviously, the heat transfern loop B PRHRS is stronger than that in loop A PRHRS.

The variation of discharge flow rate with time is shown in Fig. 18.he discharge flow rate jumps to a high value at the initial time andhen decreases to a low value. The flow rate will also oscillate in thehole process due to the pressure and water volume change.

As discussed above, when the FLB accident occurs, the integratedoop PRHRS and primary loop can establish stable natural circula-ions and the new PRHRS can provide emergency water to the SG.n addition, the innovative designed external water tank can trip

and Design 272 (2014) 45–52

the PRHRS operation. The new PRHRS can remove the core decayheat successfully and decrease the pressure of primary loop andhot leg temperature to a safe value effectively in the event of FLBaccident.

5. Conclusions

The study of a new concept PRHRS applied in CPR1000 NPP isperformed in this paper utilizing RELAP5/MOD3.4 code. The varia-tions of main transient thermal hydraulic parameters of the reactorare studied as well. Based on the analytical investigations, the fol-lowing conclusions can be obtained:

• The new PRHRS connected with the SG without valves in the loophas no effect on the steady state of the CPR1000 NPP. The free useof valve in the loop can improve the reliability of the system inthe event of accident conditions significantly and is an innovativedesign concept.

• The new PRHRS can satisfy the design requirements and removethe core residual heat successfully when SBO occurs. The externalvalve can trip the operation of the PRHRS and the stable naturalcirculations can be established in the primary loop and the PRHRS.

• In case of FLB accident, the external valve of PRHRS can trip thenew PRHRS operation to provide the emergency water to SG andremove the core residual heat successfully, realizing safety shut-down of the reactor.

Therefore, the new concept PRHRS design is successful and canmaintain the reactor safety in the event of secondary side heat sinkloss accidents. Therefore, the new PRHRS can be applied in theCPR1000 to improve the inherent safety and the concept can bealso applied in new design reactor type.

Acknowledgements

The authors would like to acknowledge the project supportedby the National Natural Science Foundation of China (grant no.11105103).

References

Achilli, A., Cattadori, G., Ferri, R., Rigamonti, M., Bianchi, F., Meloni, P.,2002.PERSEO project: experimental facility set-up and Relap5 code calculations. In:Proceedings of 2nd EMSI and 40th European Two-Phase Flow Group Meeting,Stockholm, Sweden, June 10–13. Royal Institute of Technology (KTH) (paper F3).

Ferri, R., Achilli, A., Cattadori, G., Bianchi, F., Meloni, P., 2005. Design, experi-ments and Relap5 code calculations for the PERSEO facility. Nucl. Eng. Des. 235,1201–1214.

Meloni, P., Pignatel, J.F., 1998. Theoretical design and assessment of isolationcondenser system controlled with thermal valve. In: Proceedings of 6th Interna-tional Conference on Nuclear Engineering, San Diego, USA, May 10–14, TRACK6 ICONE6-6283.

Meloni, P., 1998. Relap5/Mod3 capability to simulate the performance of isolationcondenser systems. In: Proceedings of Relap5 International User Seminar, Col-lege Station, TX, May 17–21, USNRC.

RELAP5/MOD3.3, 2001a. Code Manual, vol. 1: Code Structure, System Models, andSolution Methods, NUREG/CR-5535.

RELAP5/MOD3.3, 2001b. Code Manual Volume V: User’s Guidelines, NUREG/CR-5535.

Wang, M.J., Zhao, H., Zhang, Y.P., Su, G.H., Tian, W.X., Qiu, S.Z., 2012. Research on thedesigned emergency passive residual heat removal system during the stationblackout scenario for CPR1000. Ann. Nucl. Energy 45, 86–93.

Xiao, Z.J., Zhuo, W.B., Zheng, H., Chen, B.D., Zong, G.F., Jia, D.N., 2003. Experimentalresearch progress on passive safety systems of Chinese advanced PWR. Nucl.Eng. Des. 225, 305–313.

Zhang, Y.P., Qiu, S.Z., Su, G.H., Tian, W.X., 2011. Design and transient analyses of

emergency passive residual heat removal system of CPR1000. Part I: Air coolingcondition. Prog. Nucl. Energy 53, 471–479.

Zio, E., Pedroni, N., 2009. Estimation of the functional failure probability of athermal–hydraulic passive system by Subset Simulation. Nucl. Eng. Des. 239,580–599.