- Email: [email protected]
Conceptual design of a passive moderator cooling system for a pressure tube type natural circulation boiling water cooled reactor
Nuclear Engineering and Design 291 (2015) 261–270
Contents lists available at ScienceDirect
Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes
Conceptual design of a passive moderator cooling system for a pressure tube type natural circulation boiling water cooled reactor Mukesh Kumar a , Eshita Pal b,∗ , Arun K. Nayak a , Pallipattu K. Vijayan a a b
Reactor Engineering Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India Homi Bhabha National Institute, Anushaktinagar, Mumbai 400 094, India
h i g h l i g h t s • Passive moderator cooling system is designed to cool moderator passively during SBO. • PMCS is a system of two natural circulation loops, coupled via a heat exchanger. • RELAP5 analyses show that PMCS maintains moderator within safe limits for 7 days.
a r t i c l e
i n f o
Article history: Received 27 November 2014 Received in revised form 3 March 2015 Accepted 30 April 2015
a b s t r a c t The recent Fukushima accident has raised strong concern and apprehensions about the safety of reactors in case of a prolonged Station Black Out (SBO) continuing for several days. In view of this, a detailed study was performed simulating this condition in Advanced Heavy Water Reactor. In this study, a novel concept of moderator cooling by passive means has been introduced in the reactor design. The Passive Moderator Cooling System (PMCS) consists of a shell and tube heat exchanger designed to remove 2 MW heat from the moderator inside Calandria. The heat exchanger is located at a suitable elevation from the Calandria of the reactor, such that the hot moderator rises due to buoyancy into the heat exchanger and upon cooling from shell side water returns to Calandria forming a natural circulation loop. The shell side of the heat exchanger is also a natural circulation loop connected to an overhead large water reservoir, namely the GDWP. The objective of the PMCS is to remove the heat from the moderator in case of an SBO and maintaining its temperature below the permissible safe limit (100 ◦ C) for at least 7 days. The paper first describes the concept of the PMCS. The concept has been assessed considering a prolonged SBO for at least 7 days, through an integrated analysis performed using the code RELAP5/MOD3.2 considering all the major components of the reactor. The analysis shows that the PMCS is able to maintain the moderator temperature below boiling conditions for 7 days. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Recent Fukushima scenario has enforced the reactor designers to have a relook into the safety features of the newly designed and existing nuclear reactors. The accident has raised questions on the reliability of conventional safety systems, which are dependent on active components or system. In view of this, new reactor designs propose to adopt passive systems extensively. Passive systems use entirely passive components or active components in a limited manner (IAEA-TECDOC-1624, 2009). The driving forces in these systems are natural forces such as gravity. These are highly useful in
∗ Corresponding author. Tel.: +91 22 2559 1617; fax: +91 22 2550 5151. E-mail address: [email protected] (E. Pal). http://dx.doi.org/10.1016/j.nucengdes.2015.04.039 0029-5493/© 2015 Elsevier B.V. All rights reserved.
the conditions of non-availability of external force circulating the coolant. One of the advanced designs of nuclear reactor is the Indian Advanced Heavy Water Reactor (AHWR) (Sinha and Kakokar, 2006). The schematic of the AHWR is shown in Fig. 1A. The AHWR is an innovative reactor designed for thorium utilization with special emphasis on the use of passive systems. The AHWR is being developed at Bhabha Atomic Research Centre which will be self sustaining with 233 U being produced in the reactor core from Thorium (due to fertile to fissile isotope conversion). This is a 300 MW (e), and 920 MW(th)vertical pressure tube type, boiling water cooled, natural circulation nuclear reactor with heavy water moderator. The coolant channels are housed inside a vertical cylindrical tank called the Calandria vessel. The fuel bundle consisting of 54 fuel rods
262
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Nomenclature Notations AHWR advanced heavy water reactor CANDU Canadian deuterium uranium reactor CIESV condensate isolation emergency stop valve Calandria tube CT ECCS emergency core cooling system GDWP gravity driven water pool isolation condenser system ICS LOCA loss of coolant accident MHT main heat transport PCCS passive containment cooling system passive moderator cooling system PMCS PPIS passive poison injection system PT pressure tube SBO station black-out
is housed inside pressure tube (which carries the coolant) which is surrounded by Calandria tube, separated from the pressure tube by an annulus filled with gas (Fig. 1B; Sinha and Kakodkar, 2006). The heavy water moderator is filled inside the Calandria vessel surrounding these Calandria tubes. The fission heat from the core is removed by the natural circulation during normal operations as well as accidental scenarios. The fission heat is transferred from the fuel to the primary coolant in the Main Heat Transport system, which results in boiling of the coolant. The steam water mixture reaches a steam drum, where the steam and water separates. After separation, the steam is sent to the turbine for electricity generation and after condensation the feed water is returned back to the steam drum. The water from steam drum returns to the inlet of the pressure tubes at the bottom of the Calandria vessel, forming a complete natural circulation loop. In case of reactor shut down, the steam line to turbine is isolated and the Isolation Condensers located inside the GDWP remove the reactor decay heat. Besides, reactor design incorporates various passive safety features which include an Isolation Condenser System (ICS) for removal of decay heat, a Passive Containment Cooling System (PCCS) for containment cooling and depressurization, passive Emergency Core Cooling System (ECCS) injection directly into the channels in case of a Loss of Coolant Accident (LOCA), a Passive Poison Injection System (PPIS) to shut down the reactor passively in case of non-availability of the wired shut down system, passive submergence of the core and feeders in a sub-cooled water pool in case of a LOCA using the water from GDWP, a Passive End Shield Cooling System and a Passive Containment Isolation System (PCIS) for isolating the containment from the external atmosphere during a LOCA. Apart from these systems there are the active Calandria vault cooling system and GDWP cooling system. The recent Fukushima accident has raised strong concern and apprehensions about the safety of reactors in case of a prolonged SBO continuing for several days. In view of this, a detailed study was performed simulating such a low probable condition in AHWR. Our earlier studies have found that the reactor has the capability to remove the decay heat from reactor core for several days with the help of isolation condensers by passive means (Kumar et al., 2013). However, in that study it was observed that, a small fraction of the total reactor decay heat is transferred from the Main Heat Transport System (MHTS) to the moderator via the pressure tube and Calandria tube gap by radiation heat transfer (Table 1). Thus, the moderator temperature and pressure may rise due to non-availability of moderator cooling in prolonged SBO conditions. This rise in moderator temperature may lead to failure of rolled
joints of the Calandria tubes with the Calandria vessel end sheet, ultimately causing structural damage to the fuel assembly inside pressure tubes also. To cope with the issue and maintaining the moderator conditions within safe limits for at least 7 days, a novel concept of moderator cooling has been incorporated in the reactor design. A new passive safety system for maintaining the moderator cooling, namely the Passive Moderator Cooling System (PMCS) has been envisaged and incorporated in the reactor. The objective of the passive moderator cooling system is to remove the heat from the moderator (by radiation heat transfer across pressure tube and Calandria tube gap) in case of an SBO and maintaining its temperature below the permissible safe limit (100 ◦ C) for at least 7 days. This paper describes the concept of the PMCS. The concept has been assessed considering a prolonged SBO for at least 7 days. An integrated analysis has been performed by using the code RELAP5/MOD3.2 considering all the major components of the reactor, i.e. the main heat transport system including the core, the reactor decay heat removal system, the containment volumes, the GDWP and the passive moderator cooling system. 2. Literature survey A literature review reveals that there are a few prior studies on the possibility of a passive moderator cooling system in CANDU reactor. CANDU reactors have a large inventory of moderator which can act as a heat sink during a Loss of Coolant Accident (LOCA) (Baek and Spinks, 1994). Khartabil and Spink (1995) carried out experiments and analysis for a flashing-driven passive moderator cooling system for CANDU reactors. In this concept, the moderator exits the reactor at a temperature close to saturation so that vapor is generated in a riser connecting the Calandria to a heat exchanger. The two phase flow increases the driving force, making it possible to remove moderator heat passively. Umar et al. (1999) developed a new model of passive moderator system and tested its capability to handle LOCA accident. They predicted using CATHENA code that the moderator heat exchanger is able to remove heat for over 72 h with no saturated boiling and flow instabilities. Umar and Vecchiarelli (2000) performed a set of calculations to determine the minimum size of a moderator heat exchanger for CANDU 6 with low resistance coefficient. During LOCA, heat transferred to moderator is rejected through moderator heat exchanger that is cooled by naturally circulating water from an overhead passive emergency water system (PEWS). It may be noted that the flashing-driven passive systems are prone to flow instabilities at low powers. Dimmick et al. (2002) worked on the flashing-driven moderator cooling system and sideby-side looked into the possibility of increasing the primary coolant pressure and temperature to supercritical conditions. In the previous reported literature the Calandria vessel is a horizontal one. The horizontal Calandria tube matrix cause severe flow hindrance in case of upward directed flows of a natural circulation loop. Whereas, in the AHWR design the Calandria vessel is vertical, thus the flows are significantly different than reported literature. Also, a flashing driven system will result in high temperatures near the top end sheet of the Calandria vessel where two phase mixture exists (100 ◦ C at atmospheric pressure). This may be of concern for structural integrity of Calandria and pressure tubes. Thus a new design of the Passive Moderator Cooling system is being considered in this current work and hence it is analyzed for its performance. 3. Conceptual design of a passive moderator cooling system The Calandria Vessel of Advanced heavy Water Reactor (AHWR) is a vertical cylindrical vessel which houses 452 Calandria tubes
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
263
Fig. 1. (A) Schematic of AHWR reactor (B) structure of fuel channel.
and 62 control tubes, reactor control mechanism and heavy water as moderator. There is a constant deposition of heat in the moderator due to neutron moderation and capture, attenuation of gamma radiation as well as due to transfer of heat from the Main Heat Transport System (MHT). About 48.5 MW heat is generated in the heavy water moderator. This heat is removed continuously from the moderator by forced cooling using pumps. In the event of a prolonged station blackout (SBO), the heat generated in moderator, due to radiation heat transfer from Pressure tube to the Calandria tubes, can no longer be cooled by active means. This maximum heat produced in the moderator is estimated to be 2 MW. Due to the natural radioactive decay of the fission products in the fuel, the rate of heat generation decreases with time (Fig. 2). In turn, the heat transferred to the moderator by radiation also decreases. However, the heat transferred over time is sufficient to increase the moderator temperature to boiling conditions.
In view of this, a Passive Moderator Cooling System is designed to remove this heat, in order to prevent the pressure inside the Calandria vessel to rise beyond safe limits and to prevent boiling of moderator. Fig. 3 shows the schematic of the Passive Moderator Cooling System. The PMCS is designed to remove 2 MW heat from the moderator by means of a shell and tube type heat exchanger placed at an elevated position with respect to the Calandria. The heat generated in Calandria will increase the temperature of the moderator. This hot and less dense moderator fluid rises due to buoyant forces, enters the tube side of the shell and tube type heat exchanger. This in turn is cooled by natural circulation, by the water coming from a large overhead tank of water, namely the Gravity Driven Water Pool (GDWP). Thus forming two interdependent natural circulation loops, one between the Calandria and the heat exchanger, and the other between the heat exchanger and the GDWP.
264
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 2. Decay heat curve of the reactor AHWR (A) linear scale (B) log–log plot.
During SBO, the main moderator cooling system will not be available and Passive Moderator Cooling system removes full decay heat generated in moderator system and maintains the moderator conditions within safe limits. The PMCS is designed to operate at atmospheric pressure while maintaining the maximum moderator temperature below 100 ◦ C. 4. Performance evaluation 4.1. RELAP—Simulation of PMCS: Modelling of the system Various systems of the reactor viz Main Heat Transport System (MHTS), Isolation Condenser System (ICS), Emergency Core Cooling System (ECCS), Passive Containment Cooling System (PCCS), and Primary Containment volumes, i.e. V1 and V2 along with the Passive Moderator Cooling Systems are integrally simulated with the system code RELAP5 Mod3.2. The volume V1 and V2 are shown in Fig. 4. These are the two separate volumes contained within the primary containment volume. V1 volume encloses the Main heat transport system, during normal operation of the reactor, V1 volume is at a higher temperature of 285 ◦ C. Whereas, V2 is at a lower temperature of 40 ◦ C. Initially, steady state is obtained for MHTS and the containment volumes V1, V2 and ECCS are initialized as per the initial conditions given in Table 2. It is to be noted that the RELAP5/MOD3.2 code estimates pressure drop using friction factor formula from existing experimental correlations such as the Darcy friction factor formula. Fig. 5 shows the RELAP5 nodalization of the various systems of the reactor. All the channels in the MHTS are grouped as a single
equivalent channel having the total flow area and heat transfer area, generating the full power. This channel is connected to a single equivalent feeder of the 452 feeders of the reactor and a single equivalent tail pipe of the 452 tail pipes of the reactor. The four steam drums are combined into an equivalent steam drum. Similarly all the downcomers are modeled as an equivalent downcomer which is connected between the header and steam drum. All the tubes in the ICS are combined into an equivalent tube connected between the headers of the ICS. Five axial nodes are taken for IC tubes, which are associated with heat structure. The GDWP volume is a large volume of water. Thus, the GDWP model is divided into 3 lateral nodes and 7 axial nodes to make a good mixing model (Fig. 6). The IC is submerged in the middle section. The outlet of the PMCS shell side water flows into the top of the right section and the inlet to the shell side of the PMCS heat exchanger is taken out from the left section bottom node. The three lateral sections are interconnected. This model enhances mixing in the GDWP. In actual AHWR, there are shrouds present in the GDWP pool which cause mixing of the GDWP water. The GDWP is connected to the containment volume V2 since the GDWP is situated at the top of the containment. The two containment volumes are initially isolated from each other and filled with air at the normal operating condition as shown in Table 1. The containment volume V1 is modeled without heat absorbing capacity, while volume V2 is modeled with heat absorbing capacity. In order to estimate the heat transfer to the containment wall during condensation, the heat structure associated with the containment volume V2 is divided into 21 mesh points. The first 5 mesh points are 5 mm wide and the next 5 mesh points are 10 mm wide. This is followed
Fig. 3. Schematic of passive moderator cooling system.
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 4. Schematic of AHWR reactor cross sectional view showing volumes V1 and V2.
Fig. 5. RELAP5 nodalization of advanced heavy water reactor passive systems.
265
266
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 6. GDWP nodalization with submerged IC and inlet outlet to PMCS heat exchanger shell side.
by 5 mesh points of 25 mm width and 6 mesh points of 322.5 mm. Since concrete has poor thermal conductivity and high heat capacity, the nodalization considered in the radial direction gives a more accurate prediction of condensation on the V2 volume wall than considering an lumped one, so to account for the realistic rate of heat transfer relatively fine nodalization is taken. All the Calandria tubes are lumped in a representative one and heat transfer area of all the tubes are provided for heat removal calculations. Radiation mode of heat transfer is considered between the MHTS pressure tube and Calandria tube. Calandria tubes are connected to the moderator thermally, moderator is considered inside the Calandria vessel. All the Calandria tubes are lumped in a representative one and heat transfer area of all the tubes are provided for heat removal calculations in heat structure model of RELAP5. Calandria tubes are connected to the moderator thermally (as a heat structure). Total 24 axial nodes are considered for the Calandria tubes as well as for the Calandria vessel. The outlets to the Calandria were clubbed together into one outlet connected to the outlet header. Similar treatment was given to the inlets. The intermediate heat exchanger shell and tubes are again lumped in a single representative shell and tube. The number of nodes given to the heat exchanger is 24. The shell side of the heat exchanger is connected thermally to the tube side by heat structure defined in the input file. The heat transfer paths are shown in Fig. 7 and the heat deposition in the MHT and moderator during normal operation and SBO condition is listed in Table 1. 4.2. Scenario considered A prolonged SBO condition has been considered for analyzing the decay heat removal capability of the reactor. When the simulation starts the Passive Moderator Cooling system is assumed at a condition of rest. The power is first increased slowly to the normal
Fig. 7. Schematic heat transfer paths across various components in the AHWR systems.
operating power (920 MW) and kept constant for 7500s. After this the SBO is initiated (which is considered as time t = 0 s). With initiation of the SBO the reactor is shut down (trips). The turbine is tripped which causes closure of Condensate Isolation Emergency Stop Valve (CIESV) and the feed water supply line is also isolated at the same time due to unavailability of feed pumps. Closure of CIESV isolates the turbine from MHTS, this stops the flow of steam from the steam drum to the turbine. Thus, the MHTS becomes boxed up, which causes the pressure of the MHTS to start rising. When the pressure of the MHTS reaches the set point of the passive valve of the Isolation Condenser (IC), i.e. 76.5 bar, it opens and the ICS becomes operational. Due to unavailability of the power supply, the pneumatic pressure is lost after 30 min, which causes the active valve in the ICS to remain continuously open. If the pressure of the MHTS falls below 50 bar, ECCS injection from accumulators starts and continues until the level in the accumulators falls below 75 cm which is a low level isolation of accumulator. If the MHTS pressure falls below 3 bar then injection from GDWP may start injecting water into the core. The PMCS system is plugged in from the time SBO is initiated. The heat from the MHTS is modeled to be transferred to the moderator by radiation mode of heat transfer. The moderator temperature is initially at 67.5 ◦ C. The heat transferred from MHTS causes the moderator temperature to rise and due to buoyancy the natural circulation loop is established. Similarly, on the heat exchanger shell side the water from GDWP cools down the tube side water, and forms a natural circulation loop. This system shall maintain the
Table 1 Detailed breakdown of heat deposition in main heat transport system and moderator. System
Heat generation mode
Main heat transport system (coolant) Calandria vessel (moderator)
Fission reaction of Uranium Radioactive decay of fission product Moderation of neutrons, gamma and beta from fission Radiation heat transfer from MHTS across CT and PT gap
Heat deposition (MW) Normal operation
SBO condition (t = 0)
865 55 48.5 2
0 55 0 2
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
267
Table 2 Initial operating conditions for various systems of the reactor. System
Pressure
Temperature
Power
MHTS
70 bar
920 MW (full power)
Moderator (PMCS)
1 bar
V1 volume V2 volume Advanced Accumulator (ECCS) GDWP (ECCS) Passive valve
1.0056 bar 1.0046172 bar 55 bar 4 bar Start opening at 76.5 bar and fully opens at 79.5 bar of MHTS Opens at 79.5 bar of MHTS or 30 min after SBO (due to loss of pneumatic supply)
Core inlet = 260 ◦ C Core outlet = 285 ◦ C Calandria inlet = 65 ◦ C Calandria outlet = 70 ◦ C 285 ◦ C 40 ◦ C 40 ◦ C 40 ◦ C
Active valve
temperature of the moderator below boiling temperatures, till the time cooling water is available in GDWP. 5. Results and discussion Figs. 2 and 8 show the variation of decay power and MHTS pressure, respectively. At time t = 0 s, the reactor trips on the seismic signal. Due to a rise in pressure, the passive valves open at 76.5 bar after nearly 600 s. The steam generated in the MHT system is condensed in the ICS and this causes the pressure in the MHT to remain more or less constant. Fig. 9 shows the steam flow rate from the MHT to the ICS. The flow rate is initially higher due to larger decay heat, which reduces continuously due to reduction of decay heat with time. The active valve opens after 30 min, which helps a larger flow rate of steam into the ICS. This causes rapid condensation of steam resulting in progressive decrease of MHT pressure. At around 45 min, the MHT pressure falls to the set point of the accumulators and cold water is injected into the core passively. At about 2.5 h, accumulator injection stops due to the low level of accumulators (Fig. 10). The MHT pressure remains well above the inject set point of GDWP. As a result no injection from GDWP occurs during the whole transient. The rise in water temperature in GDWP is shown in Fig. 11. It can be seen from Fig. 11 that there is an initial surge in GDWP water temperature, which is due to the high decay heat just after reactor shutdown. It reaches the saturation temperature at atmospheric pressure after nearly 7 days. For the first 1500 s of numerical simulation, the power input to the MHT is slowly increased from zero to normal steady state
Fig. 9. Steam flow rate from MHTS to ICS.
operation power of 920 MW. From 1500 s till 9000 s, this 920 MW is kept constant in order to bring the whole system to a steady state. Uptil 9000 s, the IC is not connected to the MHTS, i.e. the steam from the MHT does not enter the IC. At 9000 s, a SBO occurs. Thus the IC is connected and the hot MHT coolant enters the GDWP, resulting in an upsurge of the GDWP temperature (Fig. 11). Also at this instant the decay heat is very high. In the later period there is a continuous steady flow of steam to the IC and also the decay
80
Pressure(bar)
70 60
Active Valve opens
50
Passive Valve opens Reactor Trip
40
Accumulator injection starts
30 20
MHT Pressure
10 0 -0.05 0.00 0.05 0.10
1
2
3
4
5
6
7
Time(days) Fig. 8. Variation of MHTS pressure.
Fig. 10. Variation of accumulator Level.
255.38 212.13 183.36 174.93 166.41 157.52 148.56 67.83 70.9 73.54 75.98 78.41 80.88 83.32 125 122 118 118 118 118 118 197.66 92.26 75.81 71.28 70.49 70.73 69.97 64.55 68.97 71.83 74.37 76.80 79.32 81.76 66.19 69.94 72.67 75.18 77.58 80.11 82.53 28.17 22.07 20.96 20.64 20.91 20.80 21.10 252.32 120.83 101.26 94.44 95.45 91.91 93.57
Average moderator Temp, Tmod (◦ C) Outlet Temp, Tout,Cal (◦ C) Inlet Temp, Tin,Cal (◦ C)
Calandria Day
Table 3 Heat balance from RELAP5 results for various components of the PMCS.
.
where mCalandria is the mass flow rate in Calandria, Cp is the specific heat for heavy water and Tout,Calandria and Tin,Calandria are the temperatures at the outlet and inlet of the Calandria. The mass flow rates, and temperature at the inlet and outlet of Calandria and shell side of heat exchanger are given in Table 3. The heat deposited and the radiation heat transfer is calculated for 7 days. It is observed that there is a difference between the radiation heat transferred and heat removed from the Calandria. This is because part of the radiation heat transferred to moderator is lost to the water in the Calandria vault surrounding the Calandria vessel. This is the heat transfer path number 4 in Fig. 7. Furthermore, heat is also lost from the shell side heat exchanger and pipes to the volume V2 which is at a lower temperature of 40 ◦ C. This is observed as a difference in the heat generated in the Calandria side and the shell side (GDWP side) of the heat exchanger. Fig. 13 shows the mass flow rate in the primary and secondary loop of the moderator circuit. It can be observed from this figure that mass flow rates are high during the initial period and decreases subsequently as decay power decreases. After 2.5 days, the moderator and the shell side of the heat exchanger mass flow rates stabilize at 10 kg/s and 21 kg/s, respectively. However, the temperature of
70.37 72.48 74.94 77.31 79.73 82.16 84.60
65.29 69.33 72.14 74.66 77.08 79.59 82.04
11.82 9.15 8.60 8.47 8.60 8.52 8.68
.
1 2 3 4 5 6 7
where ε is the emissivity, is the Stefan–Boltzmann constant, A is the heat emitting surface area (i.e. total surface area of the pressure tubes) and T is the absolute temperature (in Kelvin). The heat exchanged to the moderator is calculated using the following equation: Qmoderator = mCalandria .Cp . Tout,Calandria − Tin,Calandria = 22.3194 × 4196 × (67.5 − 47.467) = 1.876 MW
Clad Temp, Tclad (◦ C) Heat removed, QCal . (=mHX .Cp .THX ) (kW) Outlet Temp, Tout,HX (◦ C)
Inlet Temp, Tin,HX (◦ C)
= 0.5 × 5.67 × 10−8 × 903 × 545.54 − 340.5 = 1.923 MW
4
Heat removed, QCal . (=mc .Cp .TCal ) (kW)
Mass flow . rate, mC (kg/s)
4 4 Qradiation =∈ ..Apressure tube. Tpressure − TCalandria tube tube
Mass flow . rate, mHX (kg/s)
heat generated also decreases. Thus a smooth continuous increase in GDWP temperature is observed. Later when the graphs were plotted 9000 s was subtracted from the timeline in order to bring the SBO initiation at time t = 0 s. At time t = 0, SBO occurs and the reactor is shut down. Now, the only heat deposited in moderator is from radiation heat transfer across Calandria and pressure tube gap which amounts to 2 MW. The radiation heat transfer and heat balance for this time instant from the results of RELAP5 is given as follows: The MHT coolant is at an average temperature of 272.5 ◦ C. The pressure tube is also assumed to be at this temperature. The moderator is initially present at an average temperature of 67.5 ◦ C. For radiation heat transfer:
Radiation heat transfer from MHT to moderator
Fig. 11. Variation of GDWP water temperature.
Radiation heat transfer, Qrad {=εA 4 4 (Tclad − Tmod )} (kW)
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Heat exchanger shell side
268
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 12. GDWP water inventory.
the moderator entering the Calandria vessel as well as leaving the Calandria vessel (Figs. 14 and 15) increases with time as the sink (GDWP) temperature rises. In Fig. 14, for time t < 0.2 days, it is observed that the temperature at the inlet and outlet of the Calandria fluctuates. This is because the flow in the fluid of the PMCS was considered to be at rest at the beginning and with the power input (i.e. SBO initiation) at time t = 0, the system flow is getting developed under transient behavior. After 2.5 days, the temperature difference between the inlet and outlet is maintained at an almost constant value of 3 ◦ C.
Fig. 15. Variation of GDWP side coolant temperature.
In the current study it is assumed that when SBO occurs, the moderator fluid is at rest. When SBO starts, a sudden input of heat flux is given to the moderator which initiates the buoyant flows in the PMCS. The moderator is assumed to be at rest on the initiation of SBO. It takes some time for the flow to develop in the loop. During the initial phase, the temperature inside the Calandria vessel is going high as the flow is zero. When sufficient buoyancy head is generated, flow develops and temperature comes down. However, in reality the PMCS system along with the forced moderator cooling system is always online and operating during normal operation of the reactor. So 2 MW heat is always being removed by the PMCS heat exchanger and there is a steady flow in this system. Furthermore, there are flywheels connected to the pumps of the forced moderator cooling system. Consequently in the event of a SBO, the flow inside Calandria will not start from absolute zero and also the flywheels will prevent the flow from dying out fast. Thus in reality during the initial transient phase heat load, the moderator flow velocities are capable of removing the steady state maximum heat load of 2 MW. Fig. 16 shows the heat balance between the moderator primary and secondary side of the heat exchanger. It is observed that in the initial transient, the heat load to the moderator and that to the heat
Moderator Heat Load (MW) Moderator Heat loss to GDWP through HX (MW)
Fig. 13. Variation of moderator and GDWP side coolant.
269
60
50
(cntrlvar) ────── 11 Moderator Heat Load (cntrlvar) Heat loss to GDWP through HX ·············14
40
30
20
10
0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
1
2
3
4
5
6
Time (days) Fig. 14. Variation of moderator temperature.
Fig. 16. Heat balance between moderator primary and secondary side.
7
270
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
6. Conclusion A new concept of moderator cooling by passive means has been envisaged in the Advanced Heavy Water Reactor. This system was developed in light of the Fukushima accident, where active means of cooling reactor systems were unavailable. The PMCS consists of a shell and tube heat exchanger which is located at an elevation with respect to the Calandria vessel, thus forming a natural circulation loop. The shell side of the heat exchanger is connected to the GDWP forming a second natural circulation loop. The heat exchanger is designed to remove 2 MW heat which is the maximum heat generated in the event of a SBO condition. Safety analysis for a prolonged SBO condition was performed for more than 7 days. The analysis led to the conclusion that the concept works. Furthermore, the system is able to maintain the moderator temperature below boiling conditions for a time period of 7 days without any operator intervention. References Fig. 17. Variation of clad surface temperature.
exchanger are not equal. After t = 0.5 days, these heat loads equalize, i.e. the heat deposited in the moderator is entirely transferred through the heat exchanger to the GDWP. Finally, after 2 days the moderator heat load is small due to the reduction of the MHTS temperature and decay heat. Even after around 7 days, boiling is not found to occur in the GDWP. It is observed that there is still more than 7500 m3 of water in the GDWP even after 7 days of the accident (Fig. 12). During this time period the temperature in the moderator remains below the boiling conditions. The variation of clad surface temperature is shown in Fig. 17. The clad surface temperature falls below 125 ◦ C after such a long transient, which implies that these passive systems are capable of cooling the fuel and limit the moderator temperature with sufficient margins for such a prolonged period.
Baek, W.P., Spinks, N.J., 1994. CANDU passive heat rejection using the moderator. In: International Conference on New Trends in Nuclear System Thermal Hydraulics. 1, Pisa, Italy (Paper No. C22.1). Dimmick, G.R., Chatoorgoon, V., Khartabil, H.F., Duffey, R.B., 2002. Naturalconvection studies for advanced CANDU reactor concepts. Nucl. Eng. Des. 215, 27–38. Khartabil, H.F., Spink, N.J., 1995. An experimental study of a flashing-driven CANDU moderator cooling system. In: 16th Annual Conference, Canadian Nuclear Society, Saskatoon, Canada. Kumar, M., Nayak, A.K., Jain, V., Vijayan, P.K., Vaze, K.K., 2013. Managing a prolonged station blackout condition in AHWR by passive means. Nucl. Eng. Technol. 45, 605–612. IAEA TECDOC 1624, 2009. Passive Safety Systems and Natural Circulation in Water Cooled Nuclear Power Plants., pp. 1–2. Sinha, R.K., Kakokar, A., 2006. Design and development of the AHWR—the Indian thorium fuelled innovative nuclear reactor. Nucl. Eng. Des. 236, 683–700. Umar, E., Vecchiarelli, J., 2000. Parametric study of moderator heat exchanger for CANDU 6 advanced reactor. Indones. J. Nucl. Sci. Technol. 1 (1), 83–98. Umar, E., Subki, M.H., Vecchiarlli, J., 1999. Analysis of passive moderator cooling system of CANDU-6A reactor at emergency condition. In: AECL Report.
Contents lists available at ScienceDirect
Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes
Conceptual design of a passive moderator cooling system for a pressure tube type natural circulation boiling water cooled reactor Mukesh Kumar a , Eshita Pal b,∗ , Arun K. Nayak a , Pallipattu K. Vijayan a a b
Reactor Engineering Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India Homi Bhabha National Institute, Anushaktinagar, Mumbai 400 094, India
h i g h l i g h t s • Passive moderator cooling system is designed to cool moderator passively during SBO. • PMCS is a system of two natural circulation loops, coupled via a heat exchanger. • RELAP5 analyses show that PMCS maintains moderator within safe limits for 7 days.
a r t i c l e
i n f o
Article history: Received 27 November 2014 Received in revised form 3 March 2015 Accepted 30 April 2015
a b s t r a c t The recent Fukushima accident has raised strong concern and apprehensions about the safety of reactors in case of a prolonged Station Black Out (SBO) continuing for several days. In view of this, a detailed study was performed simulating this condition in Advanced Heavy Water Reactor. In this study, a novel concept of moderator cooling by passive means has been introduced in the reactor design. The Passive Moderator Cooling System (PMCS) consists of a shell and tube heat exchanger designed to remove 2 MW heat from the moderator inside Calandria. The heat exchanger is located at a suitable elevation from the Calandria of the reactor, such that the hot moderator rises due to buoyancy into the heat exchanger and upon cooling from shell side water returns to Calandria forming a natural circulation loop. The shell side of the heat exchanger is also a natural circulation loop connected to an overhead large water reservoir, namely the GDWP. The objective of the PMCS is to remove the heat from the moderator in case of an SBO and maintaining its temperature below the permissible safe limit (100 ◦ C) for at least 7 days. The paper first describes the concept of the PMCS. The concept has been assessed considering a prolonged SBO for at least 7 days, through an integrated analysis performed using the code RELAP5/MOD3.2 considering all the major components of the reactor. The analysis shows that the PMCS is able to maintain the moderator temperature below boiling conditions for 7 days. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Recent Fukushima scenario has enforced the reactor designers to have a relook into the safety features of the newly designed and existing nuclear reactors. The accident has raised questions on the reliability of conventional safety systems, which are dependent on active components or system. In view of this, new reactor designs propose to adopt passive systems extensively. Passive systems use entirely passive components or active components in a limited manner (IAEA-TECDOC-1624, 2009). The driving forces in these systems are natural forces such as gravity. These are highly useful in
∗ Corresponding author. Tel.: +91 22 2559 1617; fax: +91 22 2550 5151. E-mail address: [email protected] (E. Pal). http://dx.doi.org/10.1016/j.nucengdes.2015.04.039 0029-5493/© 2015 Elsevier B.V. All rights reserved.
the conditions of non-availability of external force circulating the coolant. One of the advanced designs of nuclear reactor is the Indian Advanced Heavy Water Reactor (AHWR) (Sinha and Kakokar, 2006). The schematic of the AHWR is shown in Fig. 1A. The AHWR is an innovative reactor designed for thorium utilization with special emphasis on the use of passive systems. The AHWR is being developed at Bhabha Atomic Research Centre which will be self sustaining with 233 U being produced in the reactor core from Thorium (due to fertile to fissile isotope conversion). This is a 300 MW (e), and 920 MW(th)vertical pressure tube type, boiling water cooled, natural circulation nuclear reactor with heavy water moderator. The coolant channels are housed inside a vertical cylindrical tank called the Calandria vessel. The fuel bundle consisting of 54 fuel rods
262
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Nomenclature Notations AHWR advanced heavy water reactor CANDU Canadian deuterium uranium reactor CIESV condensate isolation emergency stop valve Calandria tube CT ECCS emergency core cooling system GDWP gravity driven water pool isolation condenser system ICS LOCA loss of coolant accident MHT main heat transport PCCS passive containment cooling system passive moderator cooling system PMCS PPIS passive poison injection system PT pressure tube SBO station black-out
is housed inside pressure tube (which carries the coolant) which is surrounded by Calandria tube, separated from the pressure tube by an annulus filled with gas (Fig. 1B; Sinha and Kakodkar, 2006). The heavy water moderator is filled inside the Calandria vessel surrounding these Calandria tubes. The fission heat from the core is removed by the natural circulation during normal operations as well as accidental scenarios. The fission heat is transferred from the fuel to the primary coolant in the Main Heat Transport system, which results in boiling of the coolant. The steam water mixture reaches a steam drum, where the steam and water separates. After separation, the steam is sent to the turbine for electricity generation and after condensation the feed water is returned back to the steam drum. The water from steam drum returns to the inlet of the pressure tubes at the bottom of the Calandria vessel, forming a complete natural circulation loop. In case of reactor shut down, the steam line to turbine is isolated and the Isolation Condensers located inside the GDWP remove the reactor decay heat. Besides, reactor design incorporates various passive safety features which include an Isolation Condenser System (ICS) for removal of decay heat, a Passive Containment Cooling System (PCCS) for containment cooling and depressurization, passive Emergency Core Cooling System (ECCS) injection directly into the channels in case of a Loss of Coolant Accident (LOCA), a Passive Poison Injection System (PPIS) to shut down the reactor passively in case of non-availability of the wired shut down system, passive submergence of the core and feeders in a sub-cooled water pool in case of a LOCA using the water from GDWP, a Passive End Shield Cooling System and a Passive Containment Isolation System (PCIS) for isolating the containment from the external atmosphere during a LOCA. Apart from these systems there are the active Calandria vault cooling system and GDWP cooling system. The recent Fukushima accident has raised strong concern and apprehensions about the safety of reactors in case of a prolonged SBO continuing for several days. In view of this, a detailed study was performed simulating such a low probable condition in AHWR. Our earlier studies have found that the reactor has the capability to remove the decay heat from reactor core for several days with the help of isolation condensers by passive means (Kumar et al., 2013). However, in that study it was observed that, a small fraction of the total reactor decay heat is transferred from the Main Heat Transport System (MHTS) to the moderator via the pressure tube and Calandria tube gap by radiation heat transfer (Table 1). Thus, the moderator temperature and pressure may rise due to non-availability of moderator cooling in prolonged SBO conditions. This rise in moderator temperature may lead to failure of rolled
joints of the Calandria tubes with the Calandria vessel end sheet, ultimately causing structural damage to the fuel assembly inside pressure tubes also. To cope with the issue and maintaining the moderator conditions within safe limits for at least 7 days, a novel concept of moderator cooling has been incorporated in the reactor design. A new passive safety system for maintaining the moderator cooling, namely the Passive Moderator Cooling System (PMCS) has been envisaged and incorporated in the reactor. The objective of the passive moderator cooling system is to remove the heat from the moderator (by radiation heat transfer across pressure tube and Calandria tube gap) in case of an SBO and maintaining its temperature below the permissible safe limit (100 ◦ C) for at least 7 days. This paper describes the concept of the PMCS. The concept has been assessed considering a prolonged SBO for at least 7 days. An integrated analysis has been performed by using the code RELAP5/MOD3.2 considering all the major components of the reactor, i.e. the main heat transport system including the core, the reactor decay heat removal system, the containment volumes, the GDWP and the passive moderator cooling system. 2. Literature survey A literature review reveals that there are a few prior studies on the possibility of a passive moderator cooling system in CANDU reactor. CANDU reactors have a large inventory of moderator which can act as a heat sink during a Loss of Coolant Accident (LOCA) (Baek and Spinks, 1994). Khartabil and Spink (1995) carried out experiments and analysis for a flashing-driven passive moderator cooling system for CANDU reactors. In this concept, the moderator exits the reactor at a temperature close to saturation so that vapor is generated in a riser connecting the Calandria to a heat exchanger. The two phase flow increases the driving force, making it possible to remove moderator heat passively. Umar et al. (1999) developed a new model of passive moderator system and tested its capability to handle LOCA accident. They predicted using CATHENA code that the moderator heat exchanger is able to remove heat for over 72 h with no saturated boiling and flow instabilities. Umar and Vecchiarelli (2000) performed a set of calculations to determine the minimum size of a moderator heat exchanger for CANDU 6 with low resistance coefficient. During LOCA, heat transferred to moderator is rejected through moderator heat exchanger that is cooled by naturally circulating water from an overhead passive emergency water system (PEWS). It may be noted that the flashing-driven passive systems are prone to flow instabilities at low powers. Dimmick et al. (2002) worked on the flashing-driven moderator cooling system and sideby-side looked into the possibility of increasing the primary coolant pressure and temperature to supercritical conditions. In the previous reported literature the Calandria vessel is a horizontal one. The horizontal Calandria tube matrix cause severe flow hindrance in case of upward directed flows of a natural circulation loop. Whereas, in the AHWR design the Calandria vessel is vertical, thus the flows are significantly different than reported literature. Also, a flashing driven system will result in high temperatures near the top end sheet of the Calandria vessel where two phase mixture exists (100 ◦ C at atmospheric pressure). This may be of concern for structural integrity of Calandria and pressure tubes. Thus a new design of the Passive Moderator Cooling system is being considered in this current work and hence it is analyzed for its performance. 3. Conceptual design of a passive moderator cooling system The Calandria Vessel of Advanced heavy Water Reactor (AHWR) is a vertical cylindrical vessel which houses 452 Calandria tubes
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
263
Fig. 1. (A) Schematic of AHWR reactor (B) structure of fuel channel.
and 62 control tubes, reactor control mechanism and heavy water as moderator. There is a constant deposition of heat in the moderator due to neutron moderation and capture, attenuation of gamma radiation as well as due to transfer of heat from the Main Heat Transport System (MHT). About 48.5 MW heat is generated in the heavy water moderator. This heat is removed continuously from the moderator by forced cooling using pumps. In the event of a prolonged station blackout (SBO), the heat generated in moderator, due to radiation heat transfer from Pressure tube to the Calandria tubes, can no longer be cooled by active means. This maximum heat produced in the moderator is estimated to be 2 MW. Due to the natural radioactive decay of the fission products in the fuel, the rate of heat generation decreases with time (Fig. 2). In turn, the heat transferred to the moderator by radiation also decreases. However, the heat transferred over time is sufficient to increase the moderator temperature to boiling conditions.
In view of this, a Passive Moderator Cooling System is designed to remove this heat, in order to prevent the pressure inside the Calandria vessel to rise beyond safe limits and to prevent boiling of moderator. Fig. 3 shows the schematic of the Passive Moderator Cooling System. The PMCS is designed to remove 2 MW heat from the moderator by means of a shell and tube type heat exchanger placed at an elevated position with respect to the Calandria. The heat generated in Calandria will increase the temperature of the moderator. This hot and less dense moderator fluid rises due to buoyant forces, enters the tube side of the shell and tube type heat exchanger. This in turn is cooled by natural circulation, by the water coming from a large overhead tank of water, namely the Gravity Driven Water Pool (GDWP). Thus forming two interdependent natural circulation loops, one between the Calandria and the heat exchanger, and the other between the heat exchanger and the GDWP.
264
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 2. Decay heat curve of the reactor AHWR (A) linear scale (B) log–log plot.
During SBO, the main moderator cooling system will not be available and Passive Moderator Cooling system removes full decay heat generated in moderator system and maintains the moderator conditions within safe limits. The PMCS is designed to operate at atmospheric pressure while maintaining the maximum moderator temperature below 100 ◦ C. 4. Performance evaluation 4.1. RELAP—Simulation of PMCS: Modelling of the system Various systems of the reactor viz Main Heat Transport System (MHTS), Isolation Condenser System (ICS), Emergency Core Cooling System (ECCS), Passive Containment Cooling System (PCCS), and Primary Containment volumes, i.e. V1 and V2 along with the Passive Moderator Cooling Systems are integrally simulated with the system code RELAP5 Mod3.2. The volume V1 and V2 are shown in Fig. 4. These are the two separate volumes contained within the primary containment volume. V1 volume encloses the Main heat transport system, during normal operation of the reactor, V1 volume is at a higher temperature of 285 ◦ C. Whereas, V2 is at a lower temperature of 40 ◦ C. Initially, steady state is obtained for MHTS and the containment volumes V1, V2 and ECCS are initialized as per the initial conditions given in Table 2. It is to be noted that the RELAP5/MOD3.2 code estimates pressure drop using friction factor formula from existing experimental correlations such as the Darcy friction factor formula. Fig. 5 shows the RELAP5 nodalization of the various systems of the reactor. All the channels in the MHTS are grouped as a single
equivalent channel having the total flow area and heat transfer area, generating the full power. This channel is connected to a single equivalent feeder of the 452 feeders of the reactor and a single equivalent tail pipe of the 452 tail pipes of the reactor. The four steam drums are combined into an equivalent steam drum. Similarly all the downcomers are modeled as an equivalent downcomer which is connected between the header and steam drum. All the tubes in the ICS are combined into an equivalent tube connected between the headers of the ICS. Five axial nodes are taken for IC tubes, which are associated with heat structure. The GDWP volume is a large volume of water. Thus, the GDWP model is divided into 3 lateral nodes and 7 axial nodes to make a good mixing model (Fig. 6). The IC is submerged in the middle section. The outlet of the PMCS shell side water flows into the top of the right section and the inlet to the shell side of the PMCS heat exchanger is taken out from the left section bottom node. The three lateral sections are interconnected. This model enhances mixing in the GDWP. In actual AHWR, there are shrouds present in the GDWP pool which cause mixing of the GDWP water. The GDWP is connected to the containment volume V2 since the GDWP is situated at the top of the containment. The two containment volumes are initially isolated from each other and filled with air at the normal operating condition as shown in Table 1. The containment volume V1 is modeled without heat absorbing capacity, while volume V2 is modeled with heat absorbing capacity. In order to estimate the heat transfer to the containment wall during condensation, the heat structure associated with the containment volume V2 is divided into 21 mesh points. The first 5 mesh points are 5 mm wide and the next 5 mesh points are 10 mm wide. This is followed
Fig. 3. Schematic of passive moderator cooling system.
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 4. Schematic of AHWR reactor cross sectional view showing volumes V1 and V2.
Fig. 5. RELAP5 nodalization of advanced heavy water reactor passive systems.
265
266
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 6. GDWP nodalization with submerged IC and inlet outlet to PMCS heat exchanger shell side.
by 5 mesh points of 25 mm width and 6 mesh points of 322.5 mm. Since concrete has poor thermal conductivity and high heat capacity, the nodalization considered in the radial direction gives a more accurate prediction of condensation on the V2 volume wall than considering an lumped one, so to account for the realistic rate of heat transfer relatively fine nodalization is taken. All the Calandria tubes are lumped in a representative one and heat transfer area of all the tubes are provided for heat removal calculations. Radiation mode of heat transfer is considered between the MHTS pressure tube and Calandria tube. Calandria tubes are connected to the moderator thermally, moderator is considered inside the Calandria vessel. All the Calandria tubes are lumped in a representative one and heat transfer area of all the tubes are provided for heat removal calculations in heat structure model of RELAP5. Calandria tubes are connected to the moderator thermally (as a heat structure). Total 24 axial nodes are considered for the Calandria tubes as well as for the Calandria vessel. The outlets to the Calandria were clubbed together into one outlet connected to the outlet header. Similar treatment was given to the inlets. The intermediate heat exchanger shell and tubes are again lumped in a single representative shell and tube. The number of nodes given to the heat exchanger is 24. The shell side of the heat exchanger is connected thermally to the tube side by heat structure defined in the input file. The heat transfer paths are shown in Fig. 7 and the heat deposition in the MHT and moderator during normal operation and SBO condition is listed in Table 1. 4.2. Scenario considered A prolonged SBO condition has been considered for analyzing the decay heat removal capability of the reactor. When the simulation starts the Passive Moderator Cooling system is assumed at a condition of rest. The power is first increased slowly to the normal
Fig. 7. Schematic heat transfer paths across various components in the AHWR systems.
operating power (920 MW) and kept constant for 7500s. After this the SBO is initiated (which is considered as time t = 0 s). With initiation of the SBO the reactor is shut down (trips). The turbine is tripped which causes closure of Condensate Isolation Emergency Stop Valve (CIESV) and the feed water supply line is also isolated at the same time due to unavailability of feed pumps. Closure of CIESV isolates the turbine from MHTS, this stops the flow of steam from the steam drum to the turbine. Thus, the MHTS becomes boxed up, which causes the pressure of the MHTS to start rising. When the pressure of the MHTS reaches the set point of the passive valve of the Isolation Condenser (IC), i.e. 76.5 bar, it opens and the ICS becomes operational. Due to unavailability of the power supply, the pneumatic pressure is lost after 30 min, which causes the active valve in the ICS to remain continuously open. If the pressure of the MHTS falls below 50 bar, ECCS injection from accumulators starts and continues until the level in the accumulators falls below 75 cm which is a low level isolation of accumulator. If the MHTS pressure falls below 3 bar then injection from GDWP may start injecting water into the core. The PMCS system is plugged in from the time SBO is initiated. The heat from the MHTS is modeled to be transferred to the moderator by radiation mode of heat transfer. The moderator temperature is initially at 67.5 ◦ C. The heat transferred from MHTS causes the moderator temperature to rise and due to buoyancy the natural circulation loop is established. Similarly, on the heat exchanger shell side the water from GDWP cools down the tube side water, and forms a natural circulation loop. This system shall maintain the
Table 1 Detailed breakdown of heat deposition in main heat transport system and moderator. System
Heat generation mode
Main heat transport system (coolant) Calandria vessel (moderator)
Fission reaction of Uranium Radioactive decay of fission product Moderation of neutrons, gamma and beta from fission Radiation heat transfer from MHTS across CT and PT gap
Heat deposition (MW) Normal operation
SBO condition (t = 0)
865 55 48.5 2
0 55 0 2
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
267
Table 2 Initial operating conditions for various systems of the reactor. System
Pressure
Temperature
Power
MHTS
70 bar
920 MW (full power)
Moderator (PMCS)
1 bar
V1 volume V2 volume Advanced Accumulator (ECCS) GDWP (ECCS) Passive valve
1.0056 bar 1.0046172 bar 55 bar 4 bar Start opening at 76.5 bar and fully opens at 79.5 bar of MHTS Opens at 79.5 bar of MHTS or 30 min after SBO (due to loss of pneumatic supply)
Core inlet = 260 ◦ C Core outlet = 285 ◦ C Calandria inlet = 65 ◦ C Calandria outlet = 70 ◦ C 285 ◦ C 40 ◦ C 40 ◦ C 40 ◦ C
Active valve
temperature of the moderator below boiling temperatures, till the time cooling water is available in GDWP. 5. Results and discussion Figs. 2 and 8 show the variation of decay power and MHTS pressure, respectively. At time t = 0 s, the reactor trips on the seismic signal. Due to a rise in pressure, the passive valves open at 76.5 bar after nearly 600 s. The steam generated in the MHT system is condensed in the ICS and this causes the pressure in the MHT to remain more or less constant. Fig. 9 shows the steam flow rate from the MHT to the ICS. The flow rate is initially higher due to larger decay heat, which reduces continuously due to reduction of decay heat with time. The active valve opens after 30 min, which helps a larger flow rate of steam into the ICS. This causes rapid condensation of steam resulting in progressive decrease of MHT pressure. At around 45 min, the MHT pressure falls to the set point of the accumulators and cold water is injected into the core passively. At about 2.5 h, accumulator injection stops due to the low level of accumulators (Fig. 10). The MHT pressure remains well above the inject set point of GDWP. As a result no injection from GDWP occurs during the whole transient. The rise in water temperature in GDWP is shown in Fig. 11. It can be seen from Fig. 11 that there is an initial surge in GDWP water temperature, which is due to the high decay heat just after reactor shutdown. It reaches the saturation temperature at atmospheric pressure after nearly 7 days. For the first 1500 s of numerical simulation, the power input to the MHT is slowly increased from zero to normal steady state
Fig. 9. Steam flow rate from MHTS to ICS.
operation power of 920 MW. From 1500 s till 9000 s, this 920 MW is kept constant in order to bring the whole system to a steady state. Uptil 9000 s, the IC is not connected to the MHTS, i.e. the steam from the MHT does not enter the IC. At 9000 s, a SBO occurs. Thus the IC is connected and the hot MHT coolant enters the GDWP, resulting in an upsurge of the GDWP temperature (Fig. 11). Also at this instant the decay heat is very high. In the later period there is a continuous steady flow of steam to the IC and also the decay
80
Pressure(bar)
70 60
Active Valve opens
50
Passive Valve opens Reactor Trip
40
Accumulator injection starts
30 20
MHT Pressure
10 0 -0.05 0.00 0.05 0.10
1
2
3
4
5
6
7
Time(days) Fig. 8. Variation of MHTS pressure.
Fig. 10. Variation of accumulator Level.
255.38 212.13 183.36 174.93 166.41 157.52 148.56 67.83 70.9 73.54 75.98 78.41 80.88 83.32 125 122 118 118 118 118 118 197.66 92.26 75.81 71.28 70.49 70.73 69.97 64.55 68.97 71.83 74.37 76.80 79.32 81.76 66.19 69.94 72.67 75.18 77.58 80.11 82.53 28.17 22.07 20.96 20.64 20.91 20.80 21.10 252.32 120.83 101.26 94.44 95.45 91.91 93.57
Average moderator Temp, Tmod (◦ C) Outlet Temp, Tout,Cal (◦ C) Inlet Temp, Tin,Cal (◦ C)
Calandria Day
Table 3 Heat balance from RELAP5 results for various components of the PMCS.
.
where mCalandria is the mass flow rate in Calandria, Cp is the specific heat for heavy water and Tout,Calandria and Tin,Calandria are the temperatures at the outlet and inlet of the Calandria. The mass flow rates, and temperature at the inlet and outlet of Calandria and shell side of heat exchanger are given in Table 3. The heat deposited and the radiation heat transfer is calculated for 7 days. It is observed that there is a difference between the radiation heat transferred and heat removed from the Calandria. This is because part of the radiation heat transferred to moderator is lost to the water in the Calandria vault surrounding the Calandria vessel. This is the heat transfer path number 4 in Fig. 7. Furthermore, heat is also lost from the shell side heat exchanger and pipes to the volume V2 which is at a lower temperature of 40 ◦ C. This is observed as a difference in the heat generated in the Calandria side and the shell side (GDWP side) of the heat exchanger. Fig. 13 shows the mass flow rate in the primary and secondary loop of the moderator circuit. It can be observed from this figure that mass flow rates are high during the initial period and decreases subsequently as decay power decreases. After 2.5 days, the moderator and the shell side of the heat exchanger mass flow rates stabilize at 10 kg/s and 21 kg/s, respectively. However, the temperature of
70.37 72.48 74.94 77.31 79.73 82.16 84.60
65.29 69.33 72.14 74.66 77.08 79.59 82.04
11.82 9.15 8.60 8.47 8.60 8.52 8.68
.
1 2 3 4 5 6 7
where ε is the emissivity, is the Stefan–Boltzmann constant, A is the heat emitting surface area (i.e. total surface area of the pressure tubes) and T is the absolute temperature (in Kelvin). The heat exchanged to the moderator is calculated using the following equation: Qmoderator = mCalandria .Cp . Tout,Calandria − Tin,Calandria = 22.3194 × 4196 × (67.5 − 47.467) = 1.876 MW
Clad Temp, Tclad (◦ C) Heat removed, QCal . (=mHX .Cp .THX ) (kW) Outlet Temp, Tout,HX (◦ C)
Inlet Temp, Tin,HX (◦ C)
= 0.5 × 5.67 × 10−8 × 903 × 545.54 − 340.5 = 1.923 MW
4
Heat removed, QCal . (=mc .Cp .TCal ) (kW)
Mass flow . rate, mC (kg/s)
4 4 Qradiation =∈ ..Apressure tube. Tpressure − TCalandria tube tube
Mass flow . rate, mHX (kg/s)
heat generated also decreases. Thus a smooth continuous increase in GDWP temperature is observed. Later when the graphs were plotted 9000 s was subtracted from the timeline in order to bring the SBO initiation at time t = 0 s. At time t = 0, SBO occurs and the reactor is shut down. Now, the only heat deposited in moderator is from radiation heat transfer across Calandria and pressure tube gap which amounts to 2 MW. The radiation heat transfer and heat balance for this time instant from the results of RELAP5 is given as follows: The MHT coolant is at an average temperature of 272.5 ◦ C. The pressure tube is also assumed to be at this temperature. The moderator is initially present at an average temperature of 67.5 ◦ C. For radiation heat transfer:
Radiation heat transfer from MHT to moderator
Fig. 11. Variation of GDWP water temperature.
Radiation heat transfer, Qrad {=εA 4 4 (Tclad − Tmod )} (kW)
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Heat exchanger shell side
268
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
Fig. 12. GDWP water inventory.
the moderator entering the Calandria vessel as well as leaving the Calandria vessel (Figs. 14 and 15) increases with time as the sink (GDWP) temperature rises. In Fig. 14, for time t < 0.2 days, it is observed that the temperature at the inlet and outlet of the Calandria fluctuates. This is because the flow in the fluid of the PMCS was considered to be at rest at the beginning and with the power input (i.e. SBO initiation) at time t = 0, the system flow is getting developed under transient behavior. After 2.5 days, the temperature difference between the inlet and outlet is maintained at an almost constant value of 3 ◦ C.
Fig. 15. Variation of GDWP side coolant temperature.
In the current study it is assumed that when SBO occurs, the moderator fluid is at rest. When SBO starts, a sudden input of heat flux is given to the moderator which initiates the buoyant flows in the PMCS. The moderator is assumed to be at rest on the initiation of SBO. It takes some time for the flow to develop in the loop. During the initial phase, the temperature inside the Calandria vessel is going high as the flow is zero. When sufficient buoyancy head is generated, flow develops and temperature comes down. However, in reality the PMCS system along with the forced moderator cooling system is always online and operating during normal operation of the reactor. So 2 MW heat is always being removed by the PMCS heat exchanger and there is a steady flow in this system. Furthermore, there are flywheels connected to the pumps of the forced moderator cooling system. Consequently in the event of a SBO, the flow inside Calandria will not start from absolute zero and also the flywheels will prevent the flow from dying out fast. Thus in reality during the initial transient phase heat load, the moderator flow velocities are capable of removing the steady state maximum heat load of 2 MW. Fig. 16 shows the heat balance between the moderator primary and secondary side of the heat exchanger. It is observed that in the initial transient, the heat load to the moderator and that to the heat
Moderator Heat Load (MW) Moderator Heat loss to GDWP through HX (MW)
Fig. 13. Variation of moderator and GDWP side coolant.
269
60
50
(cntrlvar) ────── 11 Moderator Heat Load (cntrlvar) Heat loss to GDWP through HX ·············14
40
30
20
10
0 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
1
2
3
4
5
6
Time (days) Fig. 14. Variation of moderator temperature.
Fig. 16. Heat balance between moderator primary and secondary side.
7
270
M. Kumar et al. / Nuclear Engineering and Design 291 (2015) 261–270
6. Conclusion A new concept of moderator cooling by passive means has been envisaged in the Advanced Heavy Water Reactor. This system was developed in light of the Fukushima accident, where active means of cooling reactor systems were unavailable. The PMCS consists of a shell and tube heat exchanger which is located at an elevation with respect to the Calandria vessel, thus forming a natural circulation loop. The shell side of the heat exchanger is connected to the GDWP forming a second natural circulation loop. The heat exchanger is designed to remove 2 MW heat which is the maximum heat generated in the event of a SBO condition. Safety analysis for a prolonged SBO condition was performed for more than 7 days. The analysis led to the conclusion that the concept works. Furthermore, the system is able to maintain the moderator temperature below boiling conditions for a time period of 7 days without any operator intervention. References Fig. 17. Variation of clad surface temperature.
exchanger are not equal. After t = 0.5 days, these heat loads equalize, i.e. the heat deposited in the moderator is entirely transferred through the heat exchanger to the GDWP. Finally, after 2 days the moderator heat load is small due to the reduction of the MHTS temperature and decay heat. Even after around 7 days, boiling is not found to occur in the GDWP. It is observed that there is still more than 7500 m3 of water in the GDWP even after 7 days of the accident (Fig. 12). During this time period the temperature in the moderator remains below the boiling conditions. The variation of clad surface temperature is shown in Fig. 17. The clad surface temperature falls below 125 ◦ C after such a long transient, which implies that these passive systems are capable of cooling the fuel and limit the moderator temperature with sufficient margins for such a prolonged period.
Baek, W.P., Spinks, N.J., 1994. CANDU passive heat rejection using the moderator. In: International Conference on New Trends in Nuclear System Thermal Hydraulics. 1, Pisa, Italy (Paper No. C22.1). Dimmick, G.R., Chatoorgoon, V., Khartabil, H.F., Duffey, R.B., 2002. Naturalconvection studies for advanced CANDU reactor concepts. Nucl. Eng. Des. 215, 27–38. Khartabil, H.F., Spink, N.J., 1995. An experimental study of a flashing-driven CANDU moderator cooling system. In: 16th Annual Conference, Canadian Nuclear Society, Saskatoon, Canada. Kumar, M., Nayak, A.K., Jain, V., Vijayan, P.K., Vaze, K.K., 2013. Managing a prolonged station blackout condition in AHWR by passive means. Nucl. Eng. Technol. 45, 605–612. IAEA TECDOC 1624, 2009. Passive Safety Systems and Natural Circulation in Water Cooled Nuclear Power Plants., pp. 1–2. Sinha, R.K., Kakokar, A., 2006. Design and development of the AHWR—the Indian thorium fuelled innovative nuclear reactor. Nucl. Eng. Des. 236, 683–700. Umar, E., Vecchiarelli, J., 2000. Parametric study of moderator heat exchanger for CANDU 6 advanced reactor. Indones. J. Nucl. Sci. Technol. 1 (1), 83–98. Umar, E., Subki, M.H., Vecchiarlli, J., 1999. Analysis of passive moderator cooling system of CANDU-6A reactor at emergency condition. In: AECL Report.