© Strojni{ki vestnik 46(2000)6,370-382 © Journal of Mechanical Engineering 46(2000)6,370-382 ISSN 0039-2480 ISSN 0039-2480 UDK 621.311.25:621.039.536:621.039.58 UDC 621.311.25:621.039.536:621.039.58 Pregledni znanstveni ~lanek (1.02) Review scientific paper (1.02 ) Simuliranje odgovora zadr`evalnega hrama jedrske elektrarne med veliko izlivno nezgodo Simulation of Nuclear Power Plant Containment Response During a Large-Break Loss-of-Coolant Accident Ivo Kljenak - Borut Mavko S termohidravličnim računalniškim programom CONTAIN so bili simulirani pojavi v zadrževalnem hramu jedrske elektrarne med veliko izlivno nezgodo v dvozančnem tlačnovodnem reaktorju. Analizirani so bili tlačni in temperaturni odzivi ter porazdelitve hladiva in energije. © 2000 Strojniški vestnik. Vse pravice pridržane. (Ključne besede: elektrarne jedrske, zadrževalni hram, nezgode izlivne, simuliranje) Containment phenomena during a large-break loss-of-coolant accident in a two-loop pressurized-water reactor nuclear power plant were simulated with the CONTAIN thermal-hydraulic computer code. Pressure and temperature responses as well as coolant and energy distributions were analyzed. © 2000 Journal of Mechanical Engineering. All rights reserved. (Keywords: nuclear power plant, containment, loss-of-coolant accident, simulations) 0 UVOD Simuliranja nezgod v jedrskih elektrarnah s termohidravličnimi računalniškimi programi, ki omogočajo analizo različnih vidikov projektnih in resnih nezgod, lahko močno prispevajo k varnosti postrojenj. Čeprav vsebujejo rezultati določeno nezanesljivost, tovrstna simuliranja prispevajo tudi k boljšemu razumevanju medsebojnega vpliva fizikalnih pojavov, do katerih prihaja med nezgodami. Računalniški program CONTAIN ([1] in [2]) je bil razvit v Sandia državnih laboratorijih (ZDA) s podporo Zvezne jedrske upravne komisije ZDA. CONTAIN omogoča celovito analizo pojavov v zadrževalnem hramu jedrske elektrarne. Eden izmed osnovnih načrtovanih ciljev CONTAIN-a je upravičeno napovedovanje prehodne funkcije zadrževalnega hrama (fizikalnih, kemičnih in radioloških pogojev) med resnimi in projektnimi nezgodami. Program CONTAIN med drugim vsebuje modele za termodinamiko vodne pare in zraka, pretoke med predelki, kondenzacijo in uparjanje na konstrukcijah in aerosolih, transport, aglomeriranje in usedanje aerosolov ter zgorevanje plinov. Prav tako vsebuje modele za pojave v reaktorski votlini, kakor sta interakcija med staljeno sredico in betonom ter uparjanje hladiva v bazenu. Prevod toplote v konstrukcijah, razpadanje in transport razcepkov zaostala toplota zaradi radioaktivnih procesov ter 0 INTRODUCTION Simulations of accidents in nuclear power plants (NPPs) with thermal-hydraulic computer codes, which were developed to analyze various aspects of design-basis and severe accidents, may significantly contribute to the safety of installations. Although results carry a certain amount of uncertainty, such simulations can contribute to a better understanding of the interaction between the physical phenomena which occur during accidents. The CONTAIN computer code ([1] and [2]) was developed by Sandia National Laboratories, USA, with US Nuclear Regulatory Commission sponsorship. CONTAIN provides an integrated analysis of containment phenomena in NPPs. One of the major design objectives of CONTAIN is to provide reasonable predictions of the containment transient response (physical, chemical and radiological conditions) during the course of severe and design-basis accidents. The CONTAIN code includes atmospheric models for steam and air thermodynamics, intercompartment flows, condensation and evaporation on structures and aerosols, aerosol transport, agglomeration and deposition, and gas combustion. It also includes models for reactor cavity phenomena such as the interaction between concrete and the molten core as well as coolant-pool boiling. Heat conduction in structures, fission-product decay and transport, VH^tTPsDDIK stran 370 I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response termohidravlični in dekontaminacijski učinki varnostnih sistemov so prav tako modelirani. Pri projektnem dogodku izlivne nezgode se snov (hladivo) in energija sproščata skozi zlom cevi iz reaktorskega hladilnega sistema v zadrževalni hram. Sproščanje poteka med izlivno fazo, ponovnim polnjenjem, poplavljanje sredice in popoplavno fazo. V tej raziskavi je bil program CONTAIN (verzija 1.2) uporabljen za simuliranje odziva zadrževalnega hrama med veliko izlivno nezgodo v hladni veji dvozančnega tlačnovodnega reaktorja vrste Westinghouse. Določeni so bili tlačni in temperaturni odziv ter porazdelitev kapljevitega in parnega hladiva med prvimi 3000 s prehodnega pojava. Prav tako so analizirane energijske bilance. 1 VHODNI MODEL ZADRŽEVALNEGA HRAMA JEDRSKE ELEKTRARNE 1.1 Celice zadrževalnega hrama CONTAIN je ničrazsežni program, ki obravnava sistem zadrževalnega hrama kot mrežo medsebojno povezanih prostorov ali “celic”. V vsaki celici so navzoče tekočine mirujoče in homogene. Vsaka celica lahko pomeni resnični notranji predelek ali skupino predelkov V nekaterih primerih je primerno razdeliti predelek na več celic zaradi modeliranja pojavov, kakor so naravna konvekcija ali razslojevanje plinov. Celice so medsebojno povezane s pretokom snovi ali prevodom toplote preko vmesnih konstrukcij. CONTAIN je zasnovan za obravnavanje razmeroma majhnega števila celic. Preglednica 1. Prostornine in višine celic Table 1. Cell volumes and heights Št. No. Celica Cell 1 2 medprostor annulus 3 okolica environment 4 5 6 8 glavni predelek - prva celica main compartment - first cell radioactive-decay heating, and thermal-hydraulic and fission product decontamination effects of engineered safety features are also modeled. In a design-basis event of a loss-of-coolant accident (LOCA), mass (coolant) and energy are re-leased from the reactor coolant system (RCS) through the pipe break to the containment. These releases continue over blowdown, refill, reflood, and post-reflood phases. In this paper, the code CONTAIN (version 1.2) was used to simulate the containment response during a cold-leg large-break (LB) LOCA in a two-loop Westinghouse-type pressurized water reactor (PWR). Pressure and temperature responses, as well as coolant liquid and vapor distributions dur-ing the first 3000 s of the transient were calculated. Energy balances were also analyzed. 1 INPUT MODEL OF NUCLEAR POWER PLANT CONTAINMENT 1.1 Containment cells CONTAIN is a lumped-parameter code, which treats a containment system as a network of interconnected control volumes or “cells”. In each cell, fluids are stagnant and homogeneous. The cells represent an actual internal containment compartment or group of compartments. In some cases, a com-partment may be partitioned into several cells to model phenomena such as natural convection or gas stratification. The cells communicate with each other by means of the mass flow of material or heat con-duction through intermediate heat-transfer structures. CONTAIN is designed to use a relatively small num-ber of cells. reaktorska votlina reactor cavity predelek severnega uparjalnika north steam generator compartment predelek južnega uparjalnika south steam generator compartment predelek tlačnika pressurizer compartment glavni predelek - druga celica main compartment - second cell Prostornina Volume m3 19347 63,3 11122 60,7 1010 1000 112 13,2 Višina Height m 1135 23,1 1243 26,2 349 14,0 19347 63,3 gfin^OtJJlMlSCSD 00-6 stran 371 |^BSSITIMIGC I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response 161,00 m 159,38 m 44 5 8 4, 6 94,46 m 119,17 m 107,62 m 105,20 m 100,30 m 96,04 m 93,00 m Sl. 1. Shematični prikaz celic in tokovnih poti modela zadrževalnega hrama (višine se nanašajo samo na celice) Fig. 1. Schematic of containment model cells and flow paths (elevations refer to cells only) V tem prispevku je bil uporabljen model zadrževalnega hrama, ki temelji na [3]. Razvoj modela se je začel že v prejšnjih delih ([4] in [5]). Zadrževalni hram je razdeljen takole (slika 1, preglednica 1): - glavni predelek hrama, ki vključuje kupolo (2 celici), - kolobarjasti medprostor med jekleno stavbo in zaščitno betonsko stavbo (1 celica), - reaktorska votlina (1 celica), - predelka severnega in južnega uparjalnika (1 celica za vsak predelek), - predelek tlačnika (1 celica). Okolica je bila prav tako upoštevana kot dodatna celica za modeliranje “izgube” toplote iz zadrževalnega hrama. Opisana geometrijska razdelitev, čeprav dokaj preprosta, je zadostna za namene sedanje analize. Kakor vidimo iz primera vhodnega modela za jedrsko elektrarno Surry (ZDA) [2], so pri uporabi programa CONTAIN velike razlike med prostorninami posameznih celic običajne. 1.2 Tokovne poti med celicami zadrževalnega hrama Obravnava tokov med celicami je značilna za ničrazsežni program. Pretok med celicami poteka prek povezav, imenovanih “tokovne poti”, ki določajo In this paper, a containment model based on the model described in [3] was used. The development of the model was started in the course of ear-lier work ([4] and [5]). The containment is subdi-vided as follows (fig. 1, table 1): - containment main compartment, including the dome region (2 cells), - annulus between containment steel vessel and containment concrete-shield building (1 cell), - reactor cavity (1 cell), - north and south steam-generator compartments (1 cell for each compartment), - pressurizer compartment (1 cell). The environment was also taken into account as an additional cell to model heat “loss” from the containment. The presented geometrical configuration, although relatively simple, is sufficient for the purposes of the present analysis. As can be seen from the example input model for the Surry nuclear power plant, USA [2], large differences between the volumes of different cells are common when using the CONTAIN code. 1.2 Flow paths between containment cells The treatment of intercell flow is typical of a control-volume code. Flow is assumed to occur be-tween cells through junctions, called “flow paths”, VH^tTPsDDIK stran 372 1 7 4 I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response izmenjavo snovi in energije med celicami. Tokovne poti ne opravljajo naloge zadrževanja mase: snov, ki se pretaka po tokovni poti, se hipoma prenese v ciljno celico. Izračun masnega toka skozi tokovno pot temelji na enačbi: that determine the exchange of mass and energy be-tween the cells. The flow paths are not repositories: the material flowing into a flow path is placed im-mediately in the downstream cell. The calculation of the mass flow rate through a flow path is based on the following equation: dW dt DP-C WW (1), kjer pomenijo: W - masni tok, t - čas, DP- tlačno razliko med povezanima celicama, C - koeficient izgub, r - gostoto tekočine, A - prerez tokovne poti in L - vztrajnostno dolžino tokovne poti. Tokovne poti v vhodnem modelu so shematično prikazane na sliki 1. Njihove karakteristike so podane v preglednici 2. Vrednost koeficienta izgub C je pri vseh tokovnih poteh enaka 1,0 (glej obravnavo v [6]). 1.3 Toplotne konstrukcije v celicah hrama Ponori toplote v zadrževalnem hramu lahko prek procesov prenosa toplote in snovi zbirajo precejšnji delež toplotne energije, dovedene v zadrževalni hram med nezgodo, in tako omejujejo obremenitve hrama (tlak in temperaturo ozračja). V CONTAIN-u obstajata dve vrsti ponorov: toplotne Preglednica 2. Tokovne poti med celicami Table 2. Flow paths between cells where W is the mass flow rate, t is the time, DP is the pressure difference between connected cells, CFC is the flow-loss coefficient, r is the fluid density, A is the flow-path cross-sectional area and L is the flow-path inertial length. Flow paths in the input model are shown schematically in fig. 1. Their characteristics are tabu-lated in table 2. The flow-loss coefficient CFC is equal to 1.0 for all flow paths (see discussion in [6]). 1.3 Heat structures in containment cells Through heat and mass transfer processes, containment heat sinks can absorb a considerable fraction of the thermal energy introduced into the containment during an accident and thus provide a miti-gating effect with respect to the containment loads (atmosphere pressure and temperature). In CON- Št. No. 1 2 3 4 5 1 8 100 6 8 1 100 7 1 8 8 9 7 1 9,0 10 1 Od celice From cell 4 8 4 4 7 11 5 12 5 13 5 14 5 15 5 16 5 1 17 5 18 6 19 6 8 20 6 21 6 22 6 23 6 8 24 6 do celice to cell 1 4 8 8 1 7 8 A m2 0,31 0,0415 0,31 0,0415 1,75 0,270 0,25 0,026 2,0 1,70 1,266 9,0 1 6,17 1 0,91 1 2,36 1 0,91 1 2,35 2,25 14,66 5,64 0,91 8 2,42 8 0,91 8 2,35 2,12 15,27 A/L m 1,0 1,0 1,0 6,667 3,333 0,068 0,023 0,061 0,021 0,078 0,061 0,800 0,060 0,023 0,066 0,022 0,078 0,066 0,800 stran 373 glTMDDC I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response konstrukcije ter plasti spodnjih delov celic (v primeru TAIN, these sinks are of two main types: heat struc- obravnavanega simuliranja so to bazeni kapljevine in tures and lower cell layers (liquid pools and concrete betonska tla). Zaradi njihove pomembnosti je floors in the present simulation). Because of their modelirana vrsta procesov prenosa toplote in snovi, importance, a variety of heat and mass transfer pro- to so: naravni in prisilni konvektivni prenos toplote, cesses are modeled, such as natural and forced con- prenos toplote in snovi pri kondenzaciji, prenos vection heat transfer, condensation mass and heat toplote pri uparjanju in prevod toplote. transfer, boiling heat transfer and heat conduction. Pri izračunu konvektivnega prenosa toplote When calculating convective heat transfer z ozračja celic na toplotne konstrukcije je bilo from the cell atmosphere to heat structures, the Nusselt Nusseltovo število določeno iz naslednje povezave: number was determined from the following correlation: Nu = 0,037 Re 0,8 Pr 0,33 (2), kjer so: Nu - Nusseltovo število, Re - Reynoldsovo where Nu is the Nusselt number, Re is the Reynolds število in Pr - Prandtlovo število. Reynoldsovo number and Pr is the Prandtl number. In the CON- število je v programu CONTAIN določeno iz TAIN code, the Reynolds number is calculated from ustreznih povprečij hitrosti tokov v tokovnih poteh, appropriate averages of flow velocities in the flow priključenih na obravnavano celico. paths connected to the considered cell. V vhodnem modelu so bile upoštevane The following heat structures were taken naslednje toplotne konstrukcije [3]: valj in kupola into account in the input model [3]: containment jeklene stavbe hrama, valj in kupola zunanje betonske vessel cylinder and dome, shield building cylinder zaščitne stavbe hrama, žerjav, cevovodi, električna and dome, polar crane, piping, electrical and miscel- in različna oprema, sistemi za gretje, prezračevanje in laneous equipment, HVAC (heating, ventilation and klimatizacijo, platforme, podloge, kanal za menjavo air-conditioning) systems, platforms, embedments, goriva in notranji beton. refueling canal and interior concrete. Čeprav lahko resnične toplotne konstrukcije Although actual heat structures may as-zavzamejo dokaj zapletene oblike, se v programu sume complex shapes, they are modeled in CON-CONTAIN modelirajo kot krogle, valji ali ravne TAIN as either spheres, cylinders or slabs, which plošče, ki so lahko navpično ali vodoravno usmerjeni. may assume either vertical or horizontal orienta-Poleg tega elementi iz zgoraj naštetih kategorij niso tion. In addition, elements from the above-listed bili modelirani posamično, temveč so bili zbrani v categories were not modeled individually, but were “reprezentativne” konstrukcije. Zaradi poenostavitev grouped into “representative” structures. Due to namreč, ki so opazne pri modeliranju tokov in the inherent simplifications used in the modeling termodinamičnih procesov v ozračju hrama, of flows and atmosphere thermodynamics, a de-podrobno modeliranje toplotnih konstrukcij ne bi tailed modeling of heat structures would not nec-nujno izboljšalo zanesljivosti rezultatov simuliranja. essarily improve the reliability of the simulation Toplotne konstrukcije so bile porazdeljene results. med vsemi celicami modela zadrževalnega hrama. Heat structures were distributed among all Karakteristike toplotnih konstrukcij (material, the cells of the containment model. The characteris- geometrijska oblika, debelina) temeljijo na podatkih tics of heat structures (material, geometry, thickness) iz [3]. are based on data from [3]. 1.4 Varnostni sistemi 1.4 Engineered safety features CONTAIN omogoča tudi modeliranje CONTAIN also allows the modeling of varnostnih sistemov zadrževalnega hrama: containment engineered safety features: fan coolers ventilatorskih hladilnikov in prh. Štirje ventilatorski and containment sprays. Four fan coolers and two hladilniki in dva sistema prh so bili vključeni v vhodni spray systems were included in the input model (in model (v celicah 1 in 8). V obravnavanem scenariju cells 1 and 8). In the present simulation, it was smo predpisali, da se prhe zadrževalnega hrama prescribed that containment sprays are initiated when sprožijo, ko tlak v hramu doseže 2,06 bar, medtem the containment pressure reaches 2.06 bar, whereas ko ventilatorski hladilniki delujejo od nastanka fan coolers are assumed to be active from the time zloma. when the break occurred. Pri modeliranju prenosa toplote in snovi med When modeling heat and mass transfer kapljicami iz prh in ozračjem celic je bila uporabljena between spray droplets and cell atmosphere, the naslednja povezava za Nusseltovo število: following correlation was used for the Nusselt number: Nu = 2,0 + 0,60 Re 1/2 Pr 1/3 (3), _____00 6 SnnsjtaleJUMllfiGC] I ^BSfiTTMlliC | stran 374 i I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response medtem ko je bilo Sherwoodovo število določeno iz whereas the Sherwood number was calculated from naslednje povezave: the following correlation: Sh = 2,0 + 0,60 Re 1/2 Sc 1/3 (4), kjer sta Sh Sherwoodovo število in Sc Schmidtovo where Sh is the Sherwood number and Sc is the število. Schmidt number. 2 ZAČETNI IN ROBNI POGOJI 2 INITIAL AND BOUNDARY CONDITIONS 2.1 Začetni pogoji 2.1 Initial conditions Predpisani začetni tlak v zadrževalnem The initial pressure in the containment was hramu je bil enak 1,10 bar. Začetna izbrana prescribed to be 1.10 bar. The initial temperature of temperatura atmosfere hrama in toplotnih struktur the containment atmosphere and heat structures was je bila 322 K, razen pri reaktorski posodi, set equal to 322 K, except for the reactor pressure uparjalnikih in tlačniku, pri katerih je bila začetna vessel, steam generators and pressurizer, whose initial temperatura površine enaka 342 K. Zlom se začne surface temperature was set equal to 342 K. The break pri t = 10 s. occurs at time t = 10 s. 2.2 Robni pogoji - vir hladiva iz reaktorskega 2.2 Boundary conditions - coolant source from hladilnega sistema reactor coolant system Velika izlivna nezgoda (prvih 1000 s) je bila A LB LOCA (first 1000 s) was simulated simulirana ločeno od simuliranja s CONTAIN-om s separately from the CONTAIN simulation with the termohidravličnim računalniškim programom RELAP5/ RELAP5/MOD2 thermal-hydraulic computer code MOD2 [7]. Pri tej nezgodi pride do dvostranskega zloma [7]. A double-ended cold-leg guillotine break occurs na hladni veji med reaktorsko posodo in črpalko between the reactor vessel and the reactor coolant reaktorskega hladiva. Velikost zloma je enaka 40% pump. The break size is equal to 40% of the full cold- celotnega prereza hladne veje. Razpoložljivi so dva leg cross-section. Two high-pressure safety injection sistema za visokotlačno in en sistem za nizkotlačno systems, one low-pressure safety injection system and varnostno vbrizgavanje ter dva zbiralnika. two accumulators are available. Masna toka in specifični entalpiji hladiva Coolant (water) break mass flow rates and (vode) z obeh strani zloma so bili vključeni kot viri v specific enthalpies from both ends of the break vhodni model za CONTAIN (v celico 6). Nedavno were included as sources in the CONTAIN input so bili razviti programi, ki omogočajo sklapljanje model (in cell 6). Recently, codes which enable simuliranih procesov v reaktorskem hladilnem coupling of simulated processes in the RCS and sistemu in v zadrževalnem hramu ([8] in [9]). Pri containment have been developed ([8] and [9]). In sedanjem delu, vpliv tlaka zadrževalnega hrama na the present paper, the influence of containment tok skozi zlom iz reaktorskega hladilnega sistema ni backpressure on break flow from the RCS was not bil upoštevan. taken into account. Pri veliki izlivni nezgodi se med popoplavno During the post-reflood period of a LB fazo v reaktorskem hladilnem sistemu vzpostavi masno LOCA, an RCS mass balance exists with the ravnovesje med tokom skozi zlom in vbrizgavanjem break and emergency core-cooling system sistema za zasilno hlajenje sredice [10]. Za oba vira (ECCS) injection flow rates balanced [10]. For hladiva (z obeh strani zloma) je bil masni tok skozi zlom both coolant sources (from both ends of the med t = 1000 s in t = 3000 s izbran enak povprečni break), the break mass flow rate between t = 1000 vrednosti med t = 500 s in t = 1000 s. Energijsko s and t = 3000 s was set equal to the average value ravnovesje v reaktorskem hladilnem sistemu se between t = 500 s and t = 1000 s. A RCS energy vzpostavi med vbrizgavanjem hladne kapljevine iz balance is achieved by the injection of cold ECCS sistema za zasilno hlajenje sredice, dovajanjem toplote liquid, core heat addition, and removal of warm iz sredice in odvajanjem tople tekočine skozi zlom. Ker fluid at the break. Since the ECCS liquid is je kapljevina iz sistema za zasilno hlajenje podhlajena subcooled and the cold-leg velocities are small, in so hitrosti v hladni veji majhne, lahko predpostavimo, we may assume that steam from the core is da para iz sredice kondenzira v hladni veji ter skozi condensed within the cold leg and liquid flows zlom odteka kapljevina [10]. out of the break [10]. Rezultati simuliranja s programom According to the results of the RELAP5/ RELAP5/MOD2 kažejo, da se tlak v primarnem MOD2 simulation, the pressure in the primary system | gfin=i(gurMini5nLn 00-6_____ stran 375 I^BSSIrTMlGC I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response sistemu ustali na popoplavno vrednost 3,4 bar, ki je dosežena približno pri t = 750 s. Za oba vira je bila vrednost specifične entalpije hladiva pri t = 1000 s izbrana enaka povprečni vrednosti med t = 800 s in t = 1000 s, in entalpiji nasičene kapljevine pri 3,4 bar med t = 1500 s in t = 3000 s. Vrednosti med t = 1000 s in t = 1500 s so bile določene z linearno interpolacijo. Slika 2 prikazuje skupno maso hladiva, dovedeno iz reaktorskega hladilnega sistema s tokom skozi zlom, medtem ko slika 3 prikazuje pripadajočo skupno dovedeno entalpijo (entalpija nasičene kapljevine je definirana kot 0 J pri temperaturi 273,15 K). 7,5x105 kg settles to a post-reflood value of about 3.4 bar, which is reached at t = 750 s approximately. For both sources, the value of the coolant specific enthalpy at t = 1000 s was set equal to the average value between t = 800 s and t = 1000 s, and to the liquid saturation enthalpy at 3.4 bar between t = 1500 s and t = 3000 s. Values between t = 1000 s and t = 1500 s were calculated by linear interpolation. Figure 2 shows the cumulative coolant mass input from the RCS trough break flow, whereas figure 3 shows the corresponding cumulative enthalpy input (the liquid saturation enthalpy is defined as 0 J at 273.15 K). 5,0x105 2,5x105 0,0 0 600 1200 1800 2400 3000 t s Sl. 2. Skupna masa hladiva, dovedena s tokom skozi zlom (t - čas, m - masa) Fig. 2. Cumulative coolant mass input through break flow (t - time, m - mass) 1,0x10 12 7,5x10 11 5,0x10 11 2,5x10 11 0,0 0 600 1200 1800 2400 s 3000 Sl. 3. Skupna entalpija hladiva, dovedena s tokom skozi zlom (t - čas, H - entalpija) Fig. 3. Cumulative coolant enthalpy input through break flow (t - time, H - enthalpy) 3 REZULTATI IN RAZPRAVA Razprava v tem poglavju se nanaša na notranjost jeklene stavbe zadrževanega hrama (brez upoštevanja kolobarjastega medprostora), razen če je navedeno drugače. 3.1 Tlak v ozračju zadrževalnega hrama Slika 4 prikazuje tlak v ozračju celic zadrževalnega hrama (razlike v tlaku med različnimi celicami so zanemarljive). Takoj po nastanku zloma 3 RESULTS AND DISCUSSION Unless stated otherwise, the discussion in this section refers to the interior of the containment steel vessel, thus excluding the annulus. 3.1 Pressure in containment atmosphere Figure 4 shows the pressure of the atmosphere in the containment cells (pressure differences between different cells are negligible). VH^tTPsDDIK stran 376 m J t I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response 8x105 Pa 6x105 4x105 2x105 0 600 1200 1800 2400 3000 t s Sl. 4. Tlak v ozračju celic zadrževalnega hrama (t - čas, P - tlak) Fig. 4. Pressure in containment cells’ atmosphere (t - time, P - pressure) se tlak skoraj hipoma dvigne, kar je posledica naglega dotoka mase in energije hladiva (sl. 2, 3). Po rahlem znižanju se tlak ponovno zviša. Nihanja tlaka so posledica nihanj pretoka skozi zlom in specifične entalpije. Tlak doseže največjo vrednost 5,8 bar približno v času 1200 s, ko se začne počasi zniževati. 3.2 Temperatura ozračja zadrževalnega hrama Slika 5 prikazuje temperaturo ozračja v glavnem predelku zadrževalnega hrama (celici 1 in 8), v predelku južnega uparjalnika (celica 6) in v medprostoru (celica 2). Do najvišje temperature (z največjo vrednostjo 430 K) prihaja v celici 6, kjer je postavljen vir hladiva (tok skozi zlom iz reaktorskega hladilnega sistema). Temperatura v glavnem predelku kaže podoben vzorec obnašanja kakor tlak. Temperatura je nekoliko višja v celici 8, ki je povezana s celico 6 prek tokovnih poti. Do zvišanja temperature v medprostoru (celica 2) pride zgolj zaradi prevoda toplote skozi jekleno steno zadrževalnega hrama. 450 Immediately after the occurrence of the break, the pressure rises sharply, which is due to the sudden inflow of coolant mass and energy (see figs. 2 and 3). After a slight drop, the pressure continues to rise. Pressure oscillations are due to oscillations of break mass flow and specific enthalpy. The pressure reaches a maximum of 5.8 bar at about 1200 s, after which it starts slowly to decrease. 3.2 Temperature in containment atmosphere Figure 5 shows the atmosphere temperature in the containment main compartment (cells 1 and 8), south steam generator compartment (cell 6) and annulus (cell 2). The highest temperature (with a maximum value of 430 K) occurs in cell 6, where the coolant source (break flow from RCS) is located. The temperature in the main compartment exhibits a similar pattern of behavior to the pressure. The temperature is slightly higher in cell 8, which is connected to cell 6 through flow paths. The temperature rise in the annulus (cell 2) is caused solely by heat transfer through the containment vessel’s steel wall. K 400 350 300 0 600 1200 1800 2400 3000 ts Sl. 5. Temperatura ozračja v nekaterih celicah zadrževalnega hrama (t - čas, T - temperatura); številke označujejo celice Fig. 5. Atmosphere temperature in some containment cells (t - time, T - temperature); numbers indicate cells ^vmskmsmm 00-6 stran 377 |^BSSITIMIGC P 0 T I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response 3.3 Porazdelitev hladiva v zadrževalnem hramu Hladivo v zadrževalnem hramu izvira predvsem iz dveh virov: toka skozi zlom in prh zadrževalnega hrama Tok skozi zlom iz reaktorskega hladilnega sistema nastaja iz hladiva, navzočega pred zlomom, in hladiva, vbrizganega prek sistema za zasilno hlajenje sredice. Hladivo iz zbiralnika vode za menjavo goriva je razpoložljivo za sistem za zasilno hlajenje sredice med celotnim simulirnim prehodnim pojavom. Prhe zadrževalnega hrama črpajo hladivo iz zbiralnika vode za menjavo goriva vse do časa 1285 s, ko preidejo na obtočni način in črpajo hladivo iz zbiralnika, ki se nahaja na dnu celice 6. Kakor lahko vidimo na sliki 6, je masa hladiva, ki je dovedena s tokom skozi zlom v časovnem obdobju od 0 s do 3000 s, približno štirikrat večja od neto dovedene mase prek prh (iz zbiralnika vode za menjavo goriva). Slika 7 prikazuje porazdelitev hladiva v zadrževalnem hramu. Večina hladiva je v obliki kapljevine v bazenih na dnu celic. Pomemben delež je navzoč v ozračju v obliki pare in razpršene kapljevine, medtem ko je delež, ki se nahaja kot plast kondenzata na toplotnih konstrukcijah, za red velikosti manjši. Vendar lahko pomembnost kondenzacije pare na konstrukcijah opazimo na sliki 3.3 Coolant distribution in containment Coolant in the containment originates mainly from two sources: break flow and containment sprays. Break flow from the RCS results from coolant present before the break and coolant injected by the ECCS. Coolant from the refueling water storage tank (RWST) is available for the ECCS throughout the simulated transient. Containment sprays draw coolant from the RWST until t = 1285 s when they switch to recirculation mode and draw coolant from the sump, which is located at the bottom of cell 6. As can be seen in figure 6, the mass of the coolant input through the break flow in the time interval from 0 s to 3000 s is about four times higher than the net mass of coolant input through sprays (from RWST). Figure 7 shows the coolant distribution in the containment. Most of the coolant is located as liquid in the pools on the cell floors. A significant fraction is present in the atmosphere as vapor and dispersed liquid, whereas the fraction located as condensate film on heat structures is an order of magnitude smaller. However, the importance of vapor condensation on structures can be seen in figure 8, Sl. 6. Vir hladiva, prisotnega v zadrževalnem hramu v času t = 3000 s Fig. 6. Origin of coolant present in the containment at t = 3000 s 1,0x106 kg 7,5x105 5,0x105 2,5x105 0,0 0 600 1200 1800 2400 s 3000 Sl. 7. Masa hladiva v zadrževalnem hramu (t - čas, m - masa); a - bazeni kapljevine na dnu celic, b - para in kapljevina v ozračju celic, c - kondenzat na toplotnih strukturah Fig. 7. Coolant mass in containment (t - time, m - mass); a - liquid pools on cell floors, b - vapor and liquid in cells’ atmosphere, c - condensate on heat structures VH^tTPsDDIK stran 378 m t I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response kg 8x10 6x104 / 4x104 / 2x104 / /b H 0 0 600 1200 1800 2400 s 3000 Sl. 8. Kondenzat hladiva v zadrževalnem hramu (t - čas, m - masa); a - skupna masa kondenzata, ki je odtekel s toplotnih konstrukcij, b - preostali kondenzat na toplotnih konstrukcijah Fig. 8. Coolant condensate in containment (t - time, m - mass); a - cumulative runoff condensate from heat structures, b - remaining condensate on heat structures kg 1,0x10 7,5x104 rv \ para \^ vapor 5,0x104 ^\^ 2,5x104 kapljevina liquid 0,0 0 600 1200 1800 2400 3000 t s Sl. 9. Masa hladiva v ozračju celic zadrževalnega hrama (t - čas, m - masa) Fig. 9. Coolant mass in containment cells’ atmosphere (t - time, m - mass) 8, ki kaže, da znatna količina kondenzata odteka v bazene kapljevine. Slika 9 prikazuje porazdelitev hladiva v ozračju zadrževalnega hrama. Večina hladiva je v obliki pare. Večji del kapljevine, ki priteče v ozračje hrama z velikim masnim tokom mešanice kapljevine in pare skozi zlom med prvimi 20 s nezgode, se zelo hitro usede v bazene na dnu celic. 3.4 Porazdelitev energije v zadrževalnem hramu V času med 0 s in 3000 s je bila večina entalpije dovedena s tokom skozi zlom in le manjši delež z delovanjem prh zadrževalnega hrama (sl. 10). Glavni cilj omejitvenih ukrepov je odvajanje notranje energije iz ozračja zadrževalnega hrama ter iz celotnega hrama. Večina notranje energije je bila odvedena iz hrama z delovanjem ventilatorskih hladilnikov in le majhen del s prevodom toplote skozi jekleno steno hrama (sl. 11). Slika 12 prikazuje porazdelitev notranje energije v zadrževalnem hramu v času t = 3000 s. Večji del energije je v toplotnih konstrukcijah in v which shows that a considerable amount of condensate runs off into liquid pools. Figure 9 shows the coolant distribution in the containment atmosphere. Most of the coolant is present as vapor. Most of the liquid, which flows into the containment atmosphere with the high liquid-vapor-mixture flow rate during the first 20 s of the accident, drops very quickly into pools on the cell floors. 3.4 Energy distribution in containment Most of the enthalpy from 0 s to 3000 s was introduced via the break flow and only a small fraction by the action of containment sprays (fig. 10). The main purpose of mitigating actions is to remove internal energy, first from the containment atmosphere, and then from the entire containment. Most of the internal energy was removed from the containment by the action of fan coolers and only a small part by heat conduction through the containment vessel’s steel wall (fig. 11). Figure 12 shows the internal energy distribution in the containment at time t = 3000 s. Most of the energy is contained in the heat structures and the gfin^OtJJIMISCSD 00-6 stran 379 |^BSSIrTMlGC 4 m t 5 m I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response prhe sprays 0,26 X 1011J (3%) Sl. 10. Dovedena entalpija v zadrževalni hram od t = 0 s do t = 3000 s Fig. 10. Enthalpy input in containment from t = 0 s to t = 3000 s prenos toplote preko jeklene stene heat transfer through steel wall 11 0,05 X 1011J (2%) Sl. 11. Odvedena notranja energija iz zadrževalnega hrama od t = 0 s do t = 3000 s Fig. 11. Internal energy output from the containment from t = 0 s to t = 3000 s ozračje atmosphere \ 9% tla celic (beton) cell floors (concrete) 2% Sl. 12. Porazdelitev notranje energije v zadrževalnem hramu v času t = 3000 s Fig. 12. Internal energy distribution in the containment at t = 3000 s Sl. 13. Notranja energija, odvedena iz ozračja zadrževalnega hrama s prhami in ventilatorskimi hladilniki od t = 0 s do t = 3000 s Fig. 13. Internal energy removed from containment atmosphere by sprays and fan coolers from t = 0 s to t = 3000 s VBgfirWEBS stran 380 I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response bazenih kapljevine na dnu celic, medtem ko je precej manjši delež v ozračju hrama in betonskih tleh celic. Toplotne konstrukcije in bazeni kapljevine tako prek procesov prenosa toplote in snovi absorbirajo znaten delež dovedene toplotne energije ter na ta način omejujejo obremenitve hrama. Slika 13 ponuja primerjavo med učinki varnostnih sistemov pri odvajanju notranje energije iz ozračja zadrževalnega hrama. V obdobju med 0 s in 3000 s je bilo več ko dvakrat več energije odvedene z ventilatorskimi hladilniki kakor s prhami zadrževalnega hrama. 4 SKLEPI S termohidravličnim računalniškim programom CONTAIN so bili simulirani pojavi v zadrževalnem hramu dvozančnega tlačnovodnega reaktorja med prvimi 3000 s velike izlivne nezgode. Rezultati kažejo naslednje: 1. Tlak v ozračju zadrževalnega hrama doseže največjo vrednost 5,8 bar 1200 s po nastanku zloma. 2. Temperatura ozračja zadrževalnega hrama doseže ustrezno največjo vrednost 430 K. 3. Po začetnem izpustu iz reaktorskega hladilnega sistema je večina hladiva v zadrževalnem hramu v bazenih kapljevine, večina hladiva v ozračju pa je navzoča v obliki pare. 4. Toplotne konstrukcije in bazeni kapljevine učinkujejo omejitveno na obremenitve zadrževalnega hrama, ker na koncu simulirnega prehodnega pojava vsebujejo skoraj 90 odstotkov notranje energije. liquid pools at cell floors, whereas a much smaller fraction is contained in the containment atmosphere and the concrete cell floors. Thus, through heat- and mass-transfer processes, heat structures and liquid pools absorb a considerable fraction of the thermal energy introduced into the containment and provide a mitigating effect with respect to containment loads. Figure 13 provides a comparison between the effects of engineered safety features in removing the inter-nal energy from the containment atmosphere. In the interval from 0 s to 3000 s, more than twice as much energy was removed by fan coolers than by containment sprays. 4 CONCLUSIONS Phenomena in the containment of a two-loop pressurized water reactor during the first 3000 s of a large-break loss-of-coolant accident were simulated with the CONTAIN thermal-hydraulic computer code. The results show the following: 1. The pressure in the containment atmosphere attains a maximum value of 5.8 bar 1200 s after the occurrence of the break. 2. The temperature of the containment atmosphere attains a corresponding maximum value of 430 K. 3. After the initial release from the reactor coolant system, most of the coolant in the containment is located in liquid pools and most of the coolant in the containment atmosphere is present as vapor. 4. Heat structures and liquid pools provide a miti-gating effect with respect to containment loads as they contain almost 90% of the internal energy at the end of the simulated transient. [1] [2] [3] [4] [5] [6] [7] [8] [9] 5 LITERATURA 5 REFERENCES Murata, K.K., Carroll, D.E., Washington, K.E., Gelbard, F., Valdez, G.D., Williams, D.C., K.D. Bergeron (1989) User’s Manual for CONTAIN 1.1: A computer code for severe nuclear reactor accident containment analysis, NUREG/CR-5026, SAND87-2309, with additions C1 10O/C110P through C1 10AF, Sandia National Laboratories, Albuquerque, USA. Murata, K.K., Williams, D.C., Tills, J., Griffith, R.O., Gido, R.G., Tadios, E.L., Davis, F.J., Martinez, G.M., K.E. Washington (1997) Code Manual for CONTAIN 2.0: A computer code for nuclear reactor containment analysis, SAND97/-1735, NUREG/CR-65, Sandia National Laboratories, Albuquerque, USA. Westinghouse Energy Systems (1995) Krško nuclear power plant updated safety analysis report. Kljenak, I. (1998) Low-pressure severe accident scenario simulation with the CONTAIN code, Transactions of the American Nuclear Society, 79, 381-382, LaGrange Park, Illinois, USA. Kljenak I., Parzer I., I. Tiselj (1998) Containment phenomena during a severe accident with reactor vessel failure at low pressure in a two-loop pressurized water reactor, Proceedings of the 5th Regional Meeting “Nuclear energy in Central Europe ’98", Nuclear Society of Slovenia, Ljubljana, Slovenia. Stamps, D.W. (1998) Analyses of the thermal hydraulics in the NUPEC 1/4-scale model containment experiments, Nuclear Science and Engineering, 128, 243-269. Mavko, B., Stritar, A., A. Prošek (1993) Application of code scaling, applicability and uncertainty methodology to large-break LOCA analysis of two-loop PWR, Nuclear Engineering and Design, 143, 95-109. Smith, K.A., Baratta, A.J., G.E. Robinson (1995) Coupled RELAP5 and CONTAIN accident analysis using PVM, Nuclear Safety, 36, 1, 94-108. Kwon, Y.M., Song, J.H., S.K Lee (1997) Realistic LB-LOCA containment analysis using a merged version I. Kljenak - B. Mavko: Simuliranje odziva - Simulation of Response of RELAP5/CONTEMPT4, Proceedings, The Fifth International Topical Meeting on Nuclear Thermal Hydraulics, Operations and Safety (NUTHOS-5), Beijing, China. [10] Fletcher, C.D., R.A. Callow (1989) Long-term recovery of pressurized water reactors following a large break loss-of-coolant accident, Nuclear Engineering and Design, 110, 313-328. Naslov avtorjev: dr. Ivo Kljenak prof.dr. Borut Mavko Institut Jožef Stefan Odsek za reaktorsko tehniko Jamova 39 1000 Ljubljana Authors’ Address: Dr. Ivo Kljenak ProfDr. Borut Mavko Jožef Stefan Institute Reactor Engineering Division Jamova 39 1000 Ljubljana, Slovenia Prejeto: Received: 1.3.2000 Sprejeto: Accepted: 2.6.2000 VBgfFMK stran 382