Models for the deterministic analysis of BDBA

Models developed for the thermal-hydraulic analysis of processes in the reactor core and reactor cooling system are presented below. The models of RBMK-1500 are developed using system state-of-the-art code RELAP5 and RELAP/SCDAPSIM.

1.1 Reactor cooling system model of RBMK-1500

The RELAP5 computer code has been developed by Idaho National Engineering Laboratory [11]. It is a one-dimensional non-equilibrium two-phase thermal-hydraulic system code. The RELAP5 code has been successfully applied to PWR and BWR reactors. Since 1993 The RELAP5 model of the Ignalina NPP was used in the Lithuanian Energy Institute for the analyses of thermal-hydraulic response of the plant to various transients. The RELAP5/MOD3.2 model of the Ignalina plant (nodalization scheme) is presented in Figure 6.

The model consists of two loops. The left loop of RCS model consists of one equivalent core pass. Two drum separators are modeled as one "branch" type element (1). All downcomers are represented by a single equivalent pipe (2), further subdivided into a number of control volumes. The pump suction header (3) and the pump pressure header (8) are represented as branch objects. Three operating MCPs are represented by one equivalent element (5) with check and throttling-regulating valves. The stand-by MCP is not modeled. The bypass pipes (7) between the pump suction header and the pump pressure header is modeled with the manual valves closed. This is in agreement with a modification recently performed at the Ignalina NPP. All FCs of this left core pass are represented by an equivalent channel (12) operating at average power and coolant flow.

Compared to the model for the left loop, in the right one, the loop section between the pressure header and the DS is represented in a more detailed manner. The MCP system is modelled in more detail also (it is modelled with three equivalent pumps). The right loop model consists of three equivalent core passes. First core pass represents one single GDH with an equivalent FC of average power. Second core pass represents single GDH with failed to close check valve. A few equivalent channels of different power levels represent fuel channels, connected to this GDH. The other core pass represents the other 18 GDHs. The channels of this pass are simulated by an equivalent FC of average power. The steam separated in the separators is directed to the turbines via steam pipes (15). Two Turbine Control Valves (TCVs) organize steam supply to the turbines. The guillotine break of MCP pressure header (17) in the right loop model of RCS is modelled by a valve (18). The flow area of this valve is double of pressure header flow area. The valve (18) is connected to the volume (19), which represents the compartments covered by RCS pipelines.

image043

Fig. 6. RBMK-1500 model nodalization scheme: 1 — DS, 2 — downcomers, 3 — MCP suction header, 4 — MCP suction piping, 5 — MCPs, 6 — MCP discharge piping, 7 — bypass pipes, 8 — MCP pressure header, 9 — GDHs, 10 — lower water pipes, 11 — reactor core inlet piping, 12 — reactor core piping, 13 — reactor core outlet piping, 14 — steam-water pipes, 15 — steam pipes, 16 — check valve, 17 — single GDH, 18 — single GDH with failed to close check valve, 19 — ruptured pressure header, 20 — valve for break modeling, 21 — model of compartments, which surround the RCS pipelines

The fuel assemblies in reactor core are described as heat structure elements. The fuel channels with fuel assemblies were divided into a few (depending on the needs of modeling) equivalent groups according to the power and coolant flow rate values. For the core power of 4200 MW, the channel average power is assumed to be 2.53 MW, the maximum channel power is 3.75 MW and minimum channel power is 0.88 MW. It was assumed that approximately 95% of generated fission and decay power is generated in the fuel, and 5% in the graphite stack. More detail about the Ignalina NPP model, developed using RELAP5 code, is presented in [12, 13]. Model validation is performed by comparing calculation results and measurements using separate effect tests [14] and measurements at Ignalina NPP (integrate effects measurements) [13]. The experience of use of computer code for modeling of reactor cooling circuit in RBMK showed, that RELAP5 is very suitable for this task. One dimensional code is perfect for the modeling of thermal hydraulic and heat transfer processes in the RCS, which consists of many long pipelines without any cross flow.

For the analysis of processes, which occur in the reactor core (fuel channels of RBMK type reactors) at significant overheating of fuel assemblies up to fuel melting, specific computer tool for the analysis of processes during severe accident analysis should be used. For our purposes we used RELAP/ SCDAPSIM code [15] that is an integrated, mechanistic computer code, which models the progression of severe accidents in light-water-reactor nuclear power plants. The entire spectrum of in-vessel severe accident phenomena, including reactor — coolant-system thermal-hydraulic response, core heat up, degradation and relocation, and lower-head thermal loads, is treated in this code in a unified framework for both boiling water reactors and pressurized water reactors. Unfortunately, the RELAP/SCDAPSIM code has some limitations, related to the application for RBMK-type reactors:

• SCDAPSIM gives a possibility to define PWR or BWR fuel bundles with a user-defined fuel enrichment, but does not give a possibility to include a plant specific core content (fuel with burnable erbium absorber, which is used in RBMK-1500);

• RELAP/SCDAPSIM code does not include transport of fission products from fuel through the RCS piping and their release to confinement;

• Heat generation is defined only once for all heat structures, i. e. fuel rod, fuel channel, graphite and Control & Protection System (CPS) channels. Therefore, the consideration of heat removal by CPS channels is very complicated.

These limitations of RELAP/SCDAPSIM code were taken into account in modelling of BDBA in RBMK reactors: (1) the BWR fuel rod type was used for the modeling of RBMK fuel rods, (2) the transportation of fission products from fuel through cooling circuit was not evaluated because code limitation, (3) in order to avoid the troubles with the modeling of heat transfer from one fuel channel to another through graphite columns in radial direction in the core — the single channel model was used. Such model is acceptable for a rough analysis. The nodalization scheme of such model is presented in Figure 7. In order to perform the analysis, the following boundary conditions should be assumed for the model elements, modeled as time dependent volumes and junctions:

• pressure and water temperature in the group distribution header (3);

• pressure in the drum separators (1);

• coolant flow rate through the fuel channel with average power (2).

This model is described in more detail in the papers [16, 17] and the monograph [18].

image044

Fig. 7. Nodalization scheme of a simplified RBMK-1500 model (single fuel channel): 1 — drum separator and steam lines, 2 — fuel channel, 3 — GDH