Chris Allison Innovative Systems Software (ISS) Briar Creek Lane, Ammon, Idaho (83406), USA.

Transcription

1 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Severe Accident Management Guidelines Support Calculations for Feedwater Injection to Steam Generators during a SBO in Atucha 2 Nuclear Power Plant using RELAP5/SCDAPSIM Mod3.6 Analia Bonelli, Pablo Serrano, Oscar Mazzantini Nucleoeléctrica Argentina S.A, UG-PN, Safety Analysis and Core Design Laprida 3125 (B1603), Villa Martelli, Buenos Aires, Argentina abonelli@na-sa.com.ar, pserrano@na-sa.com.ar, mazzantini@na-sa.com.ar Chris Allison Innovative Systems Software (ISS) Briar Creek Lane, Ammon, Idaho (83406), USA. iss@cableone.net ABSTRACT For the Severe Accident Analysis of Atucha 2 Nuclear Power Plant in Argentina, a full plant nodalization for RELAP5/SCDAPSIM Mod 3.6 was developed. Atucha 2 is a PHWR, cooled and moderated by heavy water. Its 451 fuel assemblies are 6.03m high and each composed of 37 Zircaloy clad fuel rods. Each assembly is located inside an individual Zircaloy coolant channel. Heavy water coolant flows inside the channels, which are immersed inside the moderator tank. The RPV lower plenum is occupied by a massive steel structure called filling body that was designed to minimize heavy water inventory. A summary of the nodalization development is presented here: the reactor vessel model includes all the experience gained during RELAP5/SCDAPSIM Mod 3.6 code extension to represent the specific design of Atucha reactors, whereas for the primary and secondary systems and the balance of plant, advantage was taken from a previous RELAP5 full plant nodalization. One of the main uses of full plant models is Severe Accident Management Guidelines support calculations. In particular, Secondary Side Feed & Bleed has been studied. This analysis has been completed by the study of the time frame available for Steam Generator inventory recovery by turning on a feedwater pump with a dedicated diesel generator. A High Pressure Station Blackout was analyzed first, as a base case. Afterwards, Secondary Side Feed and Bleed was studied. At last, injection to both steam generators from the feedwater tank, at different times after SBO initiation was calculated. For each injection time initiation, two cases were calculated: one taking into account initial feedwater inventory only and one considering water replenishment. Main variables like cladding temperature, level in coolant channels, hydrogen production and core heatup delay have been analyzed. Calculations show that the countermeasure is in general terms equally effective, if started up to 3 hours after the initiation of the event and that a delay of around 6.hs is obtained for the cases without water replenishment to the feedwater tank. KEYWORDS SAMG, FEED AND BLEED, SEVERE ACCIDENT PROGRESSION, PHWR 1. INTRODUCTION In the frame of the thermal hydraulic simulations that lead to a severe accident required by the PSA (Probabilistic Safety Analysis) Level 2 of Atucha 2 NPP, the behaviour of the plant during the accident scenario High Pressure Station Blackout was analyzed with RELAP/SCDAP Mod3.6. The calculated 1/16

2 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety sequence takes into consideration failure of all the Diesel redundancies and all the Accumulators injection. As part of the Accident Management Program, Guides and Instructions have been developed for Atucha 2 NPP. The aim of these documents is to guide control room staff and other emergency response personnel in halting the progress of potential severe accidents and in mitigating their consequences. Severe accident management takes into account all existing plant equipment, including equipment that is not part of the standard plant safety systems. As IAEA recommends [Ref. 1], guides need to be supported by thermohydraulic calculations to justify their successfulness. This has also been requested by Argentine Regulatory Body (Autoridad Regulatoria Nuclear, ARN). In line with these recommendations, a guide to implement Feed and Bleed of the Secondary System, has been developed for Atucha 2. During its development, the possibility of the installation of a dedicated diesel generator to power at least one startup and shutdown feedwater pump (LAJ) was discussed. This accident management strategy was known to be potentially successful due to experience from results obtained from design basis accidents. The aim of this work is to confirm the successfulness of these strategies for future implementation and inclusion in the Severe Accident Management Program. One of the key features for implementing a strategy is to determine the time window in which it should be started for it to be successful. In the case of the Secondary Side Feed and Bleed, the strategy conservatively starts when batteries are depleted and Main Steam Safety Valves open. For the analyses, RELAP5/SCDAP Mod3.6 was used. This has been the code selected for Severe Accident Analysis and Severe Accident Management Strategies support calculations. 2. SEVERE ACCIDENT BEHAVIOUR CODE AND CODE DESCRIPTION RELAP/SCDAPSIM is a code designed to predict the behaviour of reactor systems during normal and accident conditions that is being developed at Innovative Systems Software (ISS) as part of the international SCDAP Development and Training Program (SDTP). RELAP/SCDAPSIM uses the publicly available SCDAP/RELAP5 [Ref. 2] models developed by the US Nuclear Regulatory Commission in combination with proprietary (a) advanced programming and numerical methods, (b) user options, and (c) models developed by ISS and other SDTP members. Even though RELAP/SCDAPSIM has been widely used for LWR applications, the unique design of Atucha 2 has presented a challenge for the code. Some of the Atucha 2 characteristics could not be correctly represented when using the commercial version RELAP/SCDAPSIM/Mod3.4. Therefore, several code extensions were introduced to RELAP/SCDAPSIM/Mod3.6. These modifications included: modeling of coolant channel to coolant channel radiation heat transfer, oxidation of the outer wall of the coolant channels, molten pool behaviour and relocation and slumping of a core with separated coolant channels (relocation to lower plenum as soon as fuel melting temperature is reached no in-core molten pool formation), and heat transfer in a lower head that includes a filling body (massive steel structure occupying most of the hemispherical volume and causing relocated debris to have a wide and thin-inheight shape). The modifications are described and assessed in [Ref.3] 3. RELAP5/SCDAP NODALIZATION FOR ATUCHA 2 Atucha 2 NPP has developed Probabilistic Safety Analyses (PSA) of Level 1 to 3. For PSA Level 1 a RELAP5 MOD3.3 model was developed, oriented to give support in probabilistic safety analysis needed in the licensing process, commissioning and support to future operation [Ref. 4]. The main components and systems of the plant were included in the nodalization, resulting in the capability to model all the expected phenomenology. For the development of this full plant nodalization, an integral platform of data and nodalization management was implemented. This platform includeed geometry and process input data, calculation 2/16

3 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety of related parameters and automatic generation of the input file for the RELAP5 code, and resulted in a great advantage in the revision of further input versions, error tracking and nodalization modifications. For the simulation of the control logics of the plant, Siemens KWU developed the DYNETZ code, a code programmed in FORTRAN, which was used during the early design stages for transient calculations. DYNETZ included, besides thermohydraulic and neutronic models, the whole control logic of the plant. As the programming capabilities of RELAP are not enough flexible to model the extensive and complex control logic of Atucha 2, a coupling with specific control subroutines in DYNETZ was developed for the above mentioned RELAP5 model of the plant [Ref.5]. Later in the project it became possible to take advantage of existing RELAP5 nodalization of the plant and develop a model suitable for calculating severe accident transients. RELAP5/SCDAPSIM was selected, since it offered the advantage that the migration from one code to the other was very straightforward. Most of the full plant thermohydraulic nodalization worked with both codes with no modifications at all. Figure 1 shows a scheme of this nodalization, which includes the Primary System, the four loops of the Moderator System, Second Heat Sink System (KAG) and Live Steam System (LBA). Main Feedwater and Startup/Shutdown Feedwater Systems are also included in the full plant nodalization, but not represented in the scheme. Fig. 1 Atucha 2 RELAP5/SCDAP Nodalization. Coupling with DYNETZ Code also allowed a very detailed representation of the Reactor Sourvillance, Limitation and Protection System and the behaviour of valves, pumps, heaters, etc. upon failure of the different electrical sources. For the migration to RELAP5/SCDAPSIM, the reactor core nodalization had to be thoroughly reviewed and adapted, to include SCDAP components in the core region which allow for the representation of core heatup, oxidation, melting and relocation, key phenomena during a Severe Accident scenario. Several heat structures were removed and replaced by SCDAP components and the channel and downcomer pipes nodalization was adapted to follow SCDAP nodalization guidelines. The pipes in the original RELAP5 model were divided into 24 axial zones: 1 for the inlet, 1 for the non-active portion, 20 for the active zone (giving boundary conditions to SCDAP components), and 2 for the outlet zone, 3/16

4 BG1 BG2 BG3 BG4 BG5 BG6 BG7 BG8 BG9 BG10 BG11 BG12 BG13 BG14 BG15 BG16 BG17 BG18 BG19 BG20 BG21 BG22 BG23 BG24 BG25 BG26 BG27 BG28 BG29 BG30 BG31 BG32 BG33 BG34 BG35 BG36 BG37 BG38 BG39 BG40 BG41 BG42 BG43 BG44 BG45 BF1 BF2 BF3 BF4 BF5 BF6 BF7 BF8 BF9 BF10 BF11 BF12 BF13 BF14 BF15 BF16 BF17 BF18 BF19 BF20 BF21 BF22 BF23 BF24 BF25 BF26 BF27 BF28 BF29 BF30 BF31 BF32 BF33 BF34 BF35 BF36 BF37 BF38 BF39 BF40 BF41 BF42 BF43 BF44 BF45 BE1 BE2 BE3 BE4 BE5 BE6 BE7 BE8 BE9 BE10 BE11 BE12 BE13 BE14 BE15 BE16 BE17 BE18 BE19 BE20 BE21 BE22 BE23 BE24 BE25 BE26 BE27 BE28 BE29 BE30 BE31 BE32 BE33 BE34 BE35 BE36 BE37 BE38 BE39 BE40 BE41 BE42 BE43 BE44 BE45 BD1 BD2 BD3 BD4 BD5 BD6 BD7 BD8 BD9 BD10 BD11 BD12 BD13 BD14 BD15 BD16 BD17 BD18 BD19 BD20 BD21 BD22 BD23 BD24 BD25 BD26 BD27 BD28 BD29 BD30 BD31 BD32 BD33 BD34 BD35 BD36 BD37 BD38 BD39 BD40 BD41 BD42 BD43 BD44 BD45 BC1 BC2 BC3 BC4 BC5 BC6 BC7 BC8 BC9 BC10 BC11 BC12 BC13 BC14 BC15 BC16 BC17 BC18 BC19 BC20 BC21 BC22 BC23 BC24 BC25 BC26 BC27 BC28 BC29 BC30 BC31 BC32 BC33 BC34 BC35 BC36 BC37 BC38 BC39 BC40 BC41 BC42 BC43 BC44 BC45 BB1 BB2 BB3 BB4 BB5 BB6 BB7 BB8 BB9 BB10 BB11 BB12 BB13 BB14 BB15 BB16 BB17 BB18 BB19 BB20 BB21 BB22 BB23 BB24 BB25 BB26 BB27 BB28 BB29 BB30 BB31 BB32 BB33 BB34 BB35 BB36 BB37 BB38 BB39 BB40 BB41 BB42 BB43 BB44 BB45 BA1 BA2 BA3 BA4 BA5 BA6 BA7 BA8 BA9 BA10 BA11 BA12 BA13 BA14 BA15 BA16 BA17 BA18 BA19 BA20 BA21 BA22 BA23 BA24 BA25 BA26 BA27 BA28 BA29 BA30 BA31 BA32 BA33 BA34 BA35 BA36 BA37 BA38 BA39 BA40 BA41 BA42 BA43 BA44 BA45 BL1 BL2 BL3 BL4 BL5 BL6 BL7 BL8 BL9 BL10 BL11 BL12 BL13 BL14 BL15 BL16 BL17 BL18 BL19 BL20 BL21 BL22 BL23 BL24 BL25 BL26 BL27 BL28 BL29 BL30 BL31 BL32 BL33 BL34 BL35 BL36 BL37 BL38 BL39 BL40 BL41 BL42 BL43 BL44 BL45 AK1 AK2 AK3 AK4 AK5 AK6 AK7 AK8 AK9 AK10 AK11 AK12 AK13 AK14 AK15 AK16 AK17 AK18 AK19 AK20 AK21 AK22 AK23 AK24 AK25 AK26 AK27 AK28 AK29 AK30 AK31 AK32 AK33 AK34 AK35 AK36 AK37 AK38 AK39 AK40 AK41 AK42 AK43 AK44 AK45 AH1 AH2 AH3 AH4 AH5 AH6 AH7 AH8 AH9 AH10 AH11 AH12 AH13 AH14 AH15 AH16 AH17 AH18 AH19 AH20 AH21 AH22 AH23 AH24 AH25 AH26 AH27 AH28 AH29 AH30 AH31 AH32 AH33 AH34 AH35 AH36 AH37 AH38 AH39 AH40 AH41 AH42 AH43 AH44 AH45 AG1 AG2 AG3 AG4 AG5 AG6 AG7 AG8 AG9 AG10 AG11 AG12 AG13 AG14 AG15 AG16 AG17 AG18 AG19 AG20 AG21 AG22 AG23 AG24 AG25 AG26 AG27 AG28 AG29 AG30 AG31 AG32 AG33 AG34 AG35 AG36 AG37 AG38 AG39 AG40 AG41 AG42 AG43 AG44 AG45 AF1 AF2 AF3 AF4 AF5 AF6 AF7 AF8 AF9 AF10 AF11 AF12 AF13 AF14 AF15 AF16 AF17 AF18 AF19 AF20 AF21 AF22 AF23 AF24 AF25 AF26 AF27 AF28 AF29 AF30 AF31 AF32 AF33 AF34 AF35 AF36 AF37 AF38 AF39 AF40 AF41 AF42 AF43 AF44 AF45 AE1 AE2 AE3 AE4 AE5 AE6 AE7 AE8 AE9 AE10 AE11 AE12 AE13 AE14 AE15 AE16 AE17 AE18 AE19 AE20 AE21 AE22 AE23 AE24 AE25 AE26 AE27 AE28 AE29 AE30 AE31 AE32 AE33 AE34 AE35 AE36 AE37 AE38 AE39 AE40 AE41 AE42 AE43 AE44 AE45 AD1 AD2 AD3 AD4 AD5 AD6 AD7 AD8 AD9 AD10 AD11 AD12 AD13 AD14 AD15 AD16 AD17 AD18 AD19 AD20 AD21 AD22 AD23 AD24 AD25 AD26 AD27 AD28 AD29 AD30 AD31 AD32 AD33 AD34 AD35 AD36 AD37 AD38 AD39 AD40 AD41 AD42 AD43 AD44 AD45 AC1 AC2 AC3 AC4 AC5 AC6 AC7 AC8 AC9 AC10 AC11 AC12 AC13 AC14 AC15 AC16 AC17 AC18 AC19 AC20 AC21 AC22 AC23 AC24 AC25 AC26 AC27 AC28 AC29 AC30 AC31 AC32 AC33 AC34 AC35 AC36 AC37 AC38 AC39 AC40 AC41 AC42 AC43 AC44 AC45 AB1 AB2 AB3 AB4 AB5 AB6 AB7 AB8 AB9 AB10 AB11 AB12 AB13 AB14 AB15 AB16 AB17 AB18 AB19 AB20 AB21 AB22 AB23 AB24 AB25 AB26 AB27 AB28 AB29 AB30 AB31 AB32 AB33 AB34 AB35 AB36 AB37 AB38 AB39 AB40 AB41 AB42 AB43 AB44 AB45 AA1 AA2 AA3 AA4 AA5 AA6 AA7 AA8 AA9 AA10 AA11 AA12 AA13 AA14 AA15 AA16 AA17 AA18 AA19 AA20 AA21 AA22 AA23 AA24 AA25 AA26 AA27 AA28 AA29 AA30 AA31 AA32 AA33 AA34 AA35 AA36 AA37 AA38 AA39 AA40 AA41 AA42 AA43 AA44 AA45 AL1 AL2 AL3 AL4 AL5 AL6 AL7 AL8 AL9 AL10 AL11 AL12 AL13 AL14 AL15 AL16 AL17 AL18 AL19 AL20 AL21 AL22 AL23 AL24 AL25 AL26 AL27 AL28 AL29 AL30 AL31 AL32 AL33 AL34 AL35 AL36 AL37 AL38 AL39 AL40 AL41 AL42 AL43 AL44 AL45 LK1 LK2 LK3 LK4 LK5 LK6 LK7 LK8 LK9 LK10 LK11 LK12 LK13 LK14 LK15 LK16 LK17 LK18 LK19 LK20 LK21 LK22 LK23 LK24 LK25 LK26 LK27 LK28 LK29 LK30 LK31 LK32 LK33 LK34 LK35 LK36 LK37 LK38 LK39 LK40 LK41 LK42 LK43 LK44 LK45 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety extended to the top of the moderator tank. This axial division was chosen to distribute in the pipe unions the located friction losses introduced by the separators of the fuel elements and was extended later to downcomer and moderator tank nodalization. This modeling takes into account that the water level in the channels and in the moderator tank may be significantly different in case of an accident. The 451 coolant channels were subdivided in 8 concentric rings [Fig 2]. The selection of the radial rings was made taking into account different aspects like the location of flow limiters, axial and radial power distribution and burn-up zones. The four inner rings correspond to thermohydraulic zone number 5 (unthrottled channels). The remaining four rings correspond to the remaining four thermohydraulic zones (1 to 4). For each ring, a representative SCDAP fuel rod component of 20 axial divisions was input associated with a shroud component used to model the fuel channel. A RELAP5 pipe type of control volume gives the boundary conditions for the fuel rods and the channel inner wall. A shroud component was also used to model moderator tank wall. Overall, 21 SCDAP components were used to model the core. SCDAP flexibility was a very important feature that allowed modeling the specific design of CNA 2 reactor core. Fig TRANSIENTS ANALYSIS BG BF BE BD BC BB BA BL AK AH AG AF AE AD AC AB AA AL LK LH LG LF LE LD LC LB LA 4.1 Base Case: High Pressure Station Blackout LH1 LH2 LH3 LH4 LH5 LH6 LH7 LH8 LH9 LH10 LH11 LH12 LH13 LH14 LH15 LH16 LH17 LH18 LH19 LH20 LH21 LH22 LH23 LH24 LH25 LH26 LH27 LH28 LH29 LH30 LH31 LH32 LH33 LH34 LH35 LH36 LH37 LH38 LH39 LH40 LH41 LH42 LH43 LH44 LH45 RING 1 LG1 LG2 LG3 LG4 LG5 LG6 LG7 LG8 LG9 LG10 LG11 LG12 LG13 LG14 LG15 LG16 LG17 LG18 LG19 LG20 LG21 LG22 LG23 LG24 LG25 LG26 LG27 LG28 LG29 LG30 LG31 LG32 LG33 LG34 LG35 LG36 LG37 LG38 LG39 LG40 LG41 LG42 LG43 LG44 LG45 RING 2 LF1 LF2 LF3 LF4 LF5 LF6 LF7 LF8 LF9 LF10 LF11 LF12 LF13 LF14 LF15 LF16 LF17 LF18 LF19 LF20 LF21 LF22 LF23 LF24 LF25 LF26 LF27 LF28 LF29 LF30 LF31 LF32 LF33 LF34 LF35 LF36 LF37 LF38 LF39 LF40 LF41 LF42 LF43 LF44 LF45 RING 3 LE1 LE2 LE3 LE4 LE5 LE6 LE7 LE8 LE9 LE10 LE11 LE12 LE13 LE14 LE15 LE16 LE17 LE18 LE19 LE20 LE21 LE22 LE23 LE24 LE25 LE26 LE27 LE28 LE29 LE30 LE31 LE32 LE33 LE34 LE35 LE36 LE37 LE38 LE39 LE40 LE41 LE42 LE43 LE44 LE45 RING 4 LD1 LD2 LD3 LD4 LD5 LD6 LD7 LD8 LD9 LD10 LD11 LD12 LD13 LD14 LD15 LD16 LD17 LD18 LD19 LD20 LD21 LD22 LD23 LD24 LD25 LD26 LD27 LD28 LD29 LD30 LD31 LD32 LD33 LD34 LD35 LD36 LD37 LD38 LD39 LD40 LD41 LD42 LD43 LD44 LD45 RING 5 LC1 LC2 LC3 LC4 LC5 LC6 LC7 LC8 LC9 LC10 LC11 LC12 LC13 LC14 LC15 LC16 LC17 LC18 LC19 LC20 LC21 LC22 LC23 LC24 LC25 LC26 LC27 LC28 LC29 LC30 LC31 LC32 LC33 LC34 LC35 LC36 LC37 LC38 LC39 LC40 LC41 LC42 LC43 LC44 LC45 RING 6 LB1 LB2 LB3 LB4 LB5 LB6 LB7 LB8 LB9 LB10 LB11 LB12 LB13 LB14 LB15 LB16 LB17 LB18 LB19 LB20 LB21 LB22 LB23 LB24 LB25 LB26 LB27 LB28 LB29 LB30 LB31 LB32 LB33 LB34 LB35 LB36 LB37 LB38 LB39 LB40 LB41 LB42 LB43 LB44 LB45 RING 7 LA1 LA2 LA3 LA4 LA5 LA6 LA7 LA8 LA9 LA10 LA11 LA12 LA13 LA14 LA15 LA16 LA17 LA18 LA19 LA20 LA21 LA22 LA23 LA24 LA25 LA26 LA27 LA28 LA29 LA30 LA31 LA32 LA33 LA34 LA35 LA36 LA37 LA38 LA39 LA40 LA41 LA42 LA43 LA44 LA45 RING Core Nodalization Channels grouping High Pressure Station Blackout scenario was selected as the base case for comparison with the Severe Accident Management Strategies herein reported. The calculated sequence takes into consideration the failure of the 4 available Diesel Redundancies, and failure of 4 Accumulator injection (operator fails to pre-set the system for injection prior to batteries depletion). Starting from normal plant operation a Station Black-Out (SBO) was assumed at time 0:00 h. The main cooling pumps and moderator pumps run down due to the loss of electrical power supply. Injection of water into the reactor cooling circuits is not available because of the loss of all pumps. Due to the loss of heat sinks both the primary pressure, as well as secondary side pressure start to increase. On the secondary side, the cooldown by a rate of 100 K/h is being initiated. This cooldown is available because the relief valves of the secondary side steam dump stations are electrically supported by batteries. The secondary side pressure decreases and consequently, the primary side pressure increase is bounded (Fig. 3 and 4/5). The water level of the steam generators is continuously decreasing due to the cooldown process and falls below 2 m at ~2000 s in both SGs (Fig. 6 and 7). Thus, the isolation of both steam valves and steam dump stations occurs. After steam isolation, small steam leakages of the 4/16

5 Pressure (bar) Pressure (bar) Pressure (bar) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety pilot valves are considered. These leakages are not strong enough to prevent a pressure increase. At about 5300 s the SG Safety Valves start to open to limit the secondary pressure. HP SBO - Base Case SS Feed&Bleed Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp Fig. 3 Primary System (Pressurizer) Pressure HP SBO - Base Case SS Feed&Bleed Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp Fig. 4 Steam Generator 1 Pressure After completion of the 100 K/h cooldown, and due to the limited heat release from the primary circuit, the primary pressure starts to increase again and the water level inside the pressurizer rises (Fig.8). At 2559s the primary pressure reaches MPa for the first time (Fig. 3). The first opening of the pressurizer safety valve occurs at 2734 s. Subsequently, the valve opens intermittently in order to limit the primary pressure. During the first cycles of the valves only steam is blown into the relief tank. Following, a water steam mixture is released into the relief tank. The valve is assumed to cycle during the whole transient calculation, i.e, it is not considered to fail. 5/16

6 SG1 Collapsed Liquid Level (m) Pressure (bar) Pressure (bar) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety HP SBO - Base Case SS Feed&Bleed Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp Fig. 5 Steam Generator 2 Pressure Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 6 Steam Generator 1 Level After 2 h, the depletion of the batteries is assumed as designed lifetime is reached. This will lead to an opening of the SG SV pilot valves and a rapid depressurization of the SG secondary side. The opening of SG SV after battery depletion is specific to CNA2 NPP design. The SG SV may reclose at low pressure due to their weight but reopen at a pressure slightly above this level. So, the SG secondary side stays depressurized for long-term. The burst disk of the relief tank fails at 8814s (~2:30 h) because of reaching a pressure gradient over the disk of 1.4 MPa. The flooding signal occurs later than 2 hours. Therefore, the water of the flooding tanks cannot be discharged into the sump due to the missing electrical power supply. 6/16

7 Pressurizer Collapsed Liquid Level (m) SG2 Collapsed Liquid Level (m) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 7 Steam Generator 2 Level Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 8 Pressurizer Collapsed Liquid Level With decreasing water level, first in the coolant channels (Fig. 9), then in the moderator tank (Fig. 10), the core starts to heat up at ~12700s under high system pressure. Simultaneously, the water level of the pressurizer slowly decreases (Fig. 8). The first radionuclides (noble gases and volatile radionuclides) are released from the fuel elements gap after bursting of the cladding at 21970s (~6 h). The delay is caused because of the numerous safety valve cycles leading to partial core cooling during the release phase by steam. Because of the exposure of the fuel elements, the core starts to heat up in the upper part. The heat-up of the lower parts of the core follows with a delay, as channels empty from top to bottom. Fuel elements temperature show a plateau between and 18000s. This is caused because radiation heat exchange occurs from the hot fuel elements to the cooler channel walls in contact with the moderator inside the moderator tank, which acts as a heat sink. When the moderator evaporates and the level in the tank decreases, the corresponding axial positions of the fuel elements and channels start to heatup. The core heat up accelerates when the oxidation process of the zircaloy cladding and zircaloy coolant channels starts at around s, when 1200ºC are reached (Fig. 11). 7/16

8 Moderator Tank Collapsed Liquid Level (m) Central Channel Collapsed Liquid Level (m) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 9 Central Channel (Ring 1) Collapsed Liquid Level Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 10 Moderator Tank Collapsed Liquid Level Cladding material starts to relocate in form of drops and agglomerates in grid spacers. As channels in Atucha are separated, the formation of an in-core molten pool is not expected. Therefore fuel element and associated coolant channel material are considered to directly relocate to the bottom as long as UO 2- Zr eutectic melting temperature is reached. Material relocation starts at about 38100s (10:30 h). Inner and upper zones of the core relocate first. The failure of the control rods before the coolant channel failure is of less importance for the CNA II core melting process compared to LWR because of the small masses and the fact that they are not in close contact with fuel rods. Therefore, even if it is possible to model such structures with SCDAP, control rods were not included in the SCDAP model presented here for simplicity. Recriticality is not a topic as reflooding of a partly destroyed core is not calculated. Heat generated during Zr oxidation reaction is of less importance compared to decay heat. Hydrogen generation occurs mainly during the core uncovering phase. Total hydrogen generation during the transient is 622kg. The calculated transient does not consider Surge Line, Hot Leg or Steam Generator 8/16

9 Maximum Core Temperature (ºC) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Tube ruptures. It is a bounding case that occurs constantly at high pressure assuming also that the PRZ safety valves cycle continuously without failing. Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 11 Maximum Core Temperature The first significant material relocation into the lower plenum happens at 40000s (11:06 h). Thereafter, the particulate debris and melt released into the lower plenum, start evaporating the remaining water. As the melt/debris spreads radially on top of the filler pieces, it gets into contact with the RPV bottom head side wall, which heats-up until RPV failure. 4.2 Secondary Side Feed & Bleed Secondary Side Feed and Bleed Strategy has been evaluated as a first approach to Accident Management in Atucha 2 NPP. The strategy can be started if the following criteria are met: level in SGs is lower than 2m (several reactor protection signals are triggered, which eases the application of the strategy), there is water available in the Feedwater Tank, which is around operation level, there is no availability of RHR System in none of the four loops. Since the plant is considered to be in a SBO state, no pumps are available. For the analysis presented here, 2hs of lifetime for Batteries has been considered. The first action consists in pressurizing the Feedwater Tank using steam from the Steam Generators via the main steam line firstly (until 2.5 bar (g)) and the auxiliary steam (pegging) line secondly, until 4.5 bar (g) are reached. This pressure corresponds to the set point of the Feedwater Tank safety valve. Following procedures for Feed and Bleed available in literature for German PWRs [Ref.6], the line communicating the SG and the Feedwater tank is closed at this point, to avoid over-pressurization of the Feedwater Tank. As a second step, the water path from the Feedwater tank to the steam generators must be open. In this analysis, only the path throught the three redundancies of the Startup and Shutdown Pumps (LAH System) has been analyzed. The setup of this water path, includes the movement of valves, many of which must be performed manually due to the lack of electricity. Special attention must also be paid to behaviour of Live Steam System valves. A detailed analysis of their behaviour was performed during PSA Level 2 project and the development of MELCOR Model [Ref. 7 and 8]. For the case of a Station Blackout transient, Main Steam Safety Valves will go to open position. This is a very specific feature of Atucha 2 and it influences the way in which SBO transients develops. Moreover, the opening of MSSV at 2 h (time considered for batteries depletion), eases the 9/16

10 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety secondary side Feed and Bleed strategy in the sense that the bleeding is automatically performed by the system without the need of operator intervention. In the case of the Isolation Valve of the MSSV, these valves can be kept fully open only if instrument air (LBX System) is available. In such a case it is considered that the secondary system pressure endstate will be around atmospheric pressure. On the contrary, if LBX system is not available, isolation valves will close when the weight of the valve more or less compensates the system pressure. This occurs at around 4 bar. This means that the valve would regulate system pressure at around 4 bar. Such a case is not considered for the Feed and Bleed, because pressure difference between Feedwater Tank and Steam Generators would be around 1.5bar, which is not a large enough driving force. This has been demonstrated by preliminary calculations. The Feed and Bleed strategy implies water flowing through stopped LAJ Pumps. These pumps constitute a high pressure loss due to friction and must be correctly taken into account. In the RELAP5 nodalization, the stopped pumps are better modelled by a single junction type of hydrodynamic component. During Atucha 2 startup, a test was performed, which was intended to measure the mass flow of Feedwater injected to Steam Generator 1 at 1 bar via one stopped Startup Feedwater Pump (only one loop) using geodesic difference and pre-presurization of the Tank as the driving force [Ref. 9]. The test was performed pre-pressurizing the Feedwater tank with air upto 3.5 bar(a) and opening the path between the Feedwater tank and the Steam Generator. Measured variables included Feedwater tank Pressure, SG Level and injected mass flow. For the determination of the stopped rotor forward friction loss coefficient (K loss), a simple model was developed to represent results obtained from the startup test [Ref. 9]. K loss was adjusted until results obtained in the startup test could be repeated. The so-obtained value for K loss was used for the full-plant calculation. The calculation was performed by setting a restart at 2hs in the base case assessed in the previous section. At the time of restart, it was considered that the Feedwater Tank was already pre-pressurized and the piping lined for the injection (valves positioned at restart). Being 2hs the design lifetime for batteries, MSSV will open due to lack of electricity as already mentioned, allowing for Feedwater injection to both Steam Generators after these are depressurized. It must be though pointed out that actions to extend lifetime of batteries during a SBO are under assessment. In such a case, since around 6hs of lifetime could be reached, the opening of MSSV will be no longer a passive but an active action requested to the operators as part of the Feed and Bleed instruction. MSSV have cycled seven times before battery depletion, when MSSV and blocking valves of the MSSV go to open position, producing a fast depressurization in the secondary system. As Instrument air is considered to be available, blocking valves of the MSSVs can be kept open and venting takes place at 1 bar. Consequently, the secondary side stays depressurized for long-term. Depressurization of the steam generators allows feedwater contained in the Feedwater Tank to flow into them. This is observed as an increase in SG level in Figures 6 and 7. The increase in the secondary side inventory acts as a recovered heat sink. Heat is transferred from the primary side to the secondary side in the steam generators (Fig. 12). The fact that heat transferred is higher than decay heat, produces a decrease in primary system temperature and pressure (Fig. 3). The system contraction can be observed as a decrease in pressurizer level in Fig. 8. Although the Primary System decrease would be low enough to allow for Accumulators injection, this has been considered to fail since Accumulators blocking valves are dependant on batteries and operator action is needed to setup the system prior to battery depletion. This operator action is not considered to take place. 10/16

11 Power (MW) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety fision + decay Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Fig. 12 Power Transferred in Steam Generators vs. Decay Heat At s, inventory in the Feedwater Tank has been totally transferred to the Steam Generators. At around s, Steam Generators dry out and heat transfer by natural circulation is strongly degraded. At first, water is evaporated in the SGs, but after dryout, the driving force for natural circulation is provided by the difference in temperature between the primary and the secondary systems. Even though that mass flow in the core is maintained at around 200kg/s up to s, heat extracted by the Steam Generators is larger than decay heat only until s. During the first phase of the accident (up to s) and from till s a free convection flow can be observed at the two upper loops of the moderator system. This flow pattern contributes to the delay in core heatup after SG dryout. For each of the lower moderator loops the flow disappears with initiation of the station black-out event, as the moderator heat exchangers are located below the core height. The primary system pressure starts to increase after SG dryout (19500.s Figure 3) while secondary system remains at around 1 bar (saturation temperature= 100.0ºC). At s a pressure of 124.5bar is reached in the primary system. Pressurizer safety valve opens some seconds later starting a new cycle of openings and closings. At this point, level in the core starts to decrease (Fig. 9). As the Pressurizer is full, water is released through the Pressurizer Relief valve during the first cycles. Later, only steam is relieved. Due to Relief Tank pressurization, the burst disk of the relief tank fails at (~10:36 h) because of reaching a pressure gradient over the disk of 1.4 MPa. Core heatup does not start until around s because even though channels are empty (Fig. 9), there is still inventory in the moderator tank, so that generated decay heat is transferred by radiation from the fuel elements to the relatively cold channel walls in contact with the moderator. When moderator tank level decreases (Fig. 10) in each of the axial positions in contact with the channel wall, cladding temperature starts to increase (Fig.11). This phenomenon, which has already been observed in the base case calculation assessed in Section 4.1, is in this case delayed around 6hs. The calculation ends at s 4.3 Feedwater Injection to Steam Generators using a Dedicated Diesel Generator During the Feed and Bleed SAMG development, the possibility of the installation of a dedicated diesel generator to power at least one startup and shutdown feedwater pump (LAJ) was discussed. This accident management strategy was known to be potentially successful due to experience from results obtained from design basis accidents. 11/16

12 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Calculations intended to confirm the successfulness of the strategy for future implementation and inclusion in the Severe Accident Management Program were performed and are presented herein. Since one of the key features for implementing a strategy is to determine the time window in which it should be started for it to be successful, injection at different timings has been evaluated, starting the LAJ pump at 1, 2 and 3 hours respectively. Another important criterion for evaluating a strategy is to know if it can be reliable in the long term. To evaluate this, another set of calculations considering water replenishment to the feedwater tank was calculated. As previously mentioned, the Isolation Valve of the MSSV can be kept open only if instrument air is available. In such a case it is considered that the system pressure will be around atmospheric pressure. On the contrary, if instrument air system is not available, isolation valves will close when the weight of the valve more or less compensates the system pressure. This occurs at around 4 bar. Taking this into consideration, two boundary conditions have been selected for the time dependent volumes representing MSSV counter pressure: cases without water replenishment to feedwater tank were calculated considering a counterpressure of 4bar, while for cases with replenishment to feedwater tank, 1 bar as counterpressure was considered. The aim of selecting different boundary conditions was to show that unlike in the case of passive Feed & Bleed, overall results are little influenced by the value taken by this boundary condition (i.e. in the availability of Instrument Air System). In the cases in which only the initial inventory of the Feedwater Tank is accounted for, injection is conservatively considered to stop when minimum level of 0.4m is reached (Figure 13). In cases in which water replenishment to Feedwater Tank is postulated, water level has been controlled between 0.4m and 2.6m. Nevertheless, the countermeasure should be equally effective regardless water level control in the tank, provided that the LAJ pump continues injection to the SGs, since LAJ are positive displacement pumps and injection will occur at a constant mass flow. Table 1 summarizes the results for the six calculated cases plus the Base Case and the Feed & Bleed strategy. In general the transient develops in the same manner as detailed in Section 4.1 (Base Case), until the time at wich injection to both steam generators by a LAJ pump powered by an external diesel generator is supposed to start. In the case of early injection at 1 h, there is a small pressure difference between the two steam generators at the start of the strategy, being SG1 dome at 53.79bar and SG2 dome at 54.09bar. This pressure difference is so that injection occurs preferably into SG1 than into SG2 as shows by SG level (Fig. 6 and 7). At the time of battery depletion (7200s), SGs domes are opened to the atmosphere and pressure between both of them is thus equalized. Injection starts then to SG2 as well. In the other cases, as batteries are already depleted and SGs opened to atmosphere, injection proceeds more or less symmetrically. For the different cases, the slow evaporation of the inventory added to the SGs, allows for a longer development of a natural convection loop in the primary system. After Steam Generator dryout, the primary system pressure starts to increase ( Figure 3) while secondary system remains at around 1 bar or 4 bar (saturation temperature= 143.6ºC) depending on the case (see Table 1). When pressure in the primary system reaches 124.5bar, pressurizer safety valve opens some seconds later, starting a new cycle of openings and closings. At this point, level in the core starts to decrease (Figure 9). Core heatup does not start until around s (for all the cases, as reported in Table 1) because even though channels are empty inside, there is still inventory in the moderator tank, so that generated decay heat is transferred by radiation from the fuel elements to the relatively cold channel walls in contact with the moderator. When moderator tank level decreases (Figure 10) in each of the axial positions in contact with the channel wall, cladding temperature starts to increase. 12/16

13 Feedwater Tank Collapsed Liquid Level (m) NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety Pump on - 1h Pump on - 1h + Rp Pump on - 2h Pump on - 2h +Rp Pump on - 3h Pump on - 3h + Rp HP SBO - Base Case SS Feed&Bleed Start of Injection Secondary Side Minimum Pressure Base Case Fig. 13 Feedwater Tank Level Table 1. Main Variable Comparison between Calculated Cases Feed &Bleed --- 2h (Battery Depletion ) Feedwater Pump Powered By Diesel Generator 1 h - Initial Inventory 1 h -With replenish ment 2 h - Initial Inventory 2 h -With replenish ment 3 h - Initial Inventory 1h 1h 2h 2h 3h 3h 3 h -With replenish ment Figure Nr bar 4 bar 1 bar 4 bar 1 bar 4 bar 1 bar Fig. 4 / Fig. 5 Injection --- Symmetri cal Minmum Level of 0.4 reached in Feedwater tank s s (2.8h) Non symetrical due to differences in pressure between the two SGs s (2.8h) Start of replenish ment Symmetrical due to pressure equalization between SGs as Batteries are depleted and SG SV opened s (3.8hs) s (3.8hs) Start of replenish ment s (4.8hs) s (4.8hs) Start of replenish ment Fig. 6 / Fig. 7 Fig. 13 SG1 Dryout 7000 s s s s s Fig. 6 SG2 Dryout 7000 s s s s s Fig. 7 End of Natural Circulation in Primary System s s s s s Start of second Openings cycle of PRZ SV Failure of Burst Disk of Relief Tank Initiation of Core Heatup End of Calculation 2734 s s s s s Fig s s no failure no failure no failure no failure s s s s s s s Fig s s s 32400s s 32400s s 32400s /16

14 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety 5. CASES COMPARISON Figure 11 shows maximum core temperature for each of the calculated transients. As can be observed, all the cases that take into consideration only water inventory initially present in feedwater tank start core heatup at around 37000s (there is a difference of s which is negligible given the time frames analyzed). Table 2 shows the total amount of heat transferred in SGs + Moderator Coolers for those three cases at s. Table 2. Total Energy Transferred to Secondary System Case Total Energy transferred in SGs+MOKs (@ s) [MJ] 1LAJ on - Injection at 3600s (initial inventory in FW TK) 6.54E+5 1LAJ on - Injection at 7200s (initial inventory in FW TK) 6.67E+5 1LAJ on - Injection at 10800s (initial inventory in FW TK) 6.86E+5 Feed&Bleed through LAJ (7200s - initial inventory in FW TK) 7.20E+5 Apparently, the effects of initiating the countermeasure are approximately the same, if it is initiated before s being that the time at which core heatup starts for the base case calculation, coincident with the time at which mass flows in the core due to natural circulation go to zero. The difference observed for the last case (passive injection), owes to the fact that injection was performed until the Feedwater Tank inventory was totally consumed, while in the previous cases (LAJ on), injection was stopped when minimum level of 0.4m was reached in Feedwater Tank. Due to the reposition of feedwater in the SGs, reflux condensation is extended. Decay heat is transferred to the secondary system and the total amount of heat transferred depends on the mass available for evaporation that in these cases is the sum of the initial inventory present in the Steam Generators plus the water added by injection via the LAJ pumps (volume initially present in the Feedwater tank + piping). Nevertheless, there are other variables that influence the overall behaviour of the system. Extracting decay heat while primary and secondary systems are at a higher pressure is considered to be more efficient, since heat of evaporation is lower at higher pressures. That means that the same amount of heat extracted from the primary system evaporates a smaller mass of water in the secondary. Furthermore, inventory loss to the containment must be analyzed. For Cases in which LAJ pump is started at 1 h after SBO initiation, injection starts before Pressurizer SV opening. For the following cases, the later the countermeasure is initiated, the more cycles have been suffered by the PRZ-SV (Figure 3), leading to a larger loss of primary inventory to the containment. The implementation of the strategy (only taking into consideration inventory initially in the SGs plus that in the Feedwater tank), delays core heatup around 6 hours. For the three cases with water replenishment to the Feedwater tank, temperatures can be maintained low for a larger period of time. It is important to note that these cases were calculated taking into consideration a lower pressure in the Steam Generators. This means that heat transfer occurs with the secondary system at a lower temperature. Nevertheless, Figure 12 shows little difference of the extracted heat in the short term period (when comparing cases with injection at the same time and with/without water replenishment). Figure 6 and 7 show water level in SG1 and SG2 respectively. For Cases in which level in both SGs increases considerably, heat transfer in the SGs is degraded, which is evidenced by a temperature increase (Figure 11). Therefore, the countermeasure must be accompanied by a constant surveillance in the SGs level. That means that instrumentation intended for that, must be available during the whole Severe Accident Scenario and designed to withstand the conditions expected during the whole countermeasure. 14/16

15 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety 6. CONCLUSIONS Accident Management Strategy Station Blackout with Secondary System Feed & Bleed via LAJ pumps has been assessed. Seven different cases have been studied: Feed & Bleed using geodesic difference and pre-presurization of Feedwater Tank is compared with active Feedwater injection to SGs powering a Startup and Shudown pump with a dedicated diesel generator, at different times, considering and without considering water replenishment to the tank. The transients start from the plant operating at 100% rated power. Failure of external power sources and four diesel generators have been considered. Accumulator injection is also supposed to fail, as was considered for the base case calculation. Overall results show that the countermeasure results equally effective if it is initiated before s being that the time at which core heatup starts for the base case calculation. This suggests that the most important variable is the total inventory of water injected to the Steam Generators while injection mass flow has a minor influence. Core heatup is delayed around 6 hours. NOMENCLATURE ARN CNA2 FW TK IAEA ISS LAJ LWR MSSV PSA PRZ PWR RPV SAM SAMG SBO SG SV Autoridad Regulatoria Nuclear Central Nuclear Atucha 2 Presidente Nestor Carlos Kirchner Feedwater Tank International Atomic Energy Agency Innovative Systems Software Startup and Shutdown Feedwater Pumps Light Water Reactor Main Steam Safety Valve Probabilistic Safety Asessment Pressurizer Pressurized Water Reactor Reactor Pressure Vessel Severe Accident Management Severe Accident Management Guidelines Station Blackout Steam Generator Safety Valves ACKNOWLEDGMENTS To Larry Siefken, who was in charge of the code development and was always open to questions and fruitful technical discussions. Lessons learned from him remain with us. REFERENCES 1. Implementation of Accident Management Programs in Nuclear Power Plants - Safety Report Series Nr.32, IAEA, Vienna, SCDAP/RELAP5/MOD3.2 Code Manual, NUREG/CR Rev 1. INEL-96/0422, October, BONELLI, A., SIEFKEN, L. et al. Summary of Severe Accident Assessment for Atucha 2 Nuclear Power Plant using RELAP5/SCDAP Mod3.6, NUTHOS , Japan, Dec /16

16 NUTHOS-11: The 11th International Topical Meeting on Nuclear Reactor Thermal Hydraulics, Operation and Safety 4. González, J., Zanocco, P. et al. RELAP Model for Atucha II PHWR - ICONE Proceedings of the 17th International Conference on Nuclear Engineering - July 12-16, 2009, Brussels, Belgium 5. THELER, G. On the implementation of reactor control, limitation and protection systems as Fortran routines and data exchange mechanisms with other codes TECNA Technical Report N-IT14-102, Buenos Aires, 6/8/ ROTH-SEEFRID, H.; FEIGEL, A.; MOSER, H.J. Implementation of bleed and feed procedures in Siemens PWRs, Nuclear Engineering and Design, 148 (1994). 7. SONNENKALB, M.; STEINRÖTTER, T.H. Probabilistic Safety Analysis (PSA) Level 2 for CNAII CNAII Melcor Input Description, Technical Report GRS 01/2012, Cologne, Germany, September LÖFFLER, H.; SONNENKALB, M. Probabilistic Safety Analysis (PSA) Level 2 for CNAII Assessment of Specific Phenomena of Severe Accidents, Technical Report GRS 07/2012, Cologne, Germany, February BONELLI, A. Station Blackout in Atucha 2 with one LAJ pump injection to Steam Generators as Severe Accident Management Strategy - Internal Technical Report Number NA-BN Buenos Aires, July /16