THERMOHYDRAULIC RESEARCH FOR THE CORE OF THE BREST-OD-300 REACTOR

Transcription

1 11th International Conference on Nuclear Engineering Tokyo, Japan, April 20-23, 2003 ICONE THERMOHYDRAULIC RESEARCH FOR THE CORE OF THE BREST-OD-300 REACTOR V.P. Smirnov FSUE RDIPE, Mosco, Russia A.I. Filin FSUE RDIPE, Mosco, Russia A.G. Sila-Novitsky FSUE RDIPE, Mosco, Russia V.N. Leonov FSUE RDIPE, Mosco, Russia A.V. Zhukov A.D. Efanov A.P. Sorokin J.A. Kuzina ABSTRACT The results of experimental and computational thermohydraulic research for the experimental demonstration lead-cooled fast reactor BREST-OD-300 are considered. The generalized recommendations (formula, graphic dependence) to estimate the Nusselt numbers and temperature non-uniformities along a perimeter of fuel pins of this reactor for a number of cases representing the greatest practical interest for thermohydraulic substantiation of the reactor core are given. 1. PURPOSE OF RESEARCHES The conceptual studies of lead-cooled fast reactors have shon that this direction is perspective from a point of vie of design of reactors hich have inherent advanced safety features [1, 2]. It as necessary to carry on experimental researches of thermohydraulics for the core. Taking into account a lo level of heat-transfer coefficients of the lead coolant in contrast to sodium (BN-type reactors) and practically unexplored square rod arrangement used in these reactors, it as necessary to investigate ho heat-transfer coefficients depend on the Peclet number (Pe), the rod pitch (s/d), spacer grids and other factors hich are characteristic for the BREST-type reactor. The large attention as given to study of temperature non-uniformities of pins in the regular lattice, and pins located at the boundary of zones ith different diameters and energy releases of pins. 2. ORGANIZATION OF RESEARCH AND MODELING Experimental studies of heat-transfer coefficients and temperature fields of fuel rods for the BREST-type reactors have been carried out using three thermohydraulic models hich have identical structures and are only distinguished by pitches (s/d=1.46; 1.28 and 1.25) and also availability or absence of spacer grids. The models are the bundles of 25 model fuel rods ith square arrangement of rods located into the rectangular cover (Figure 1, a, b). Along the central model fuel rod, hich is rotary, the surface temperature measurements ere conducted along perimeter and length of the model fuel rod by microthermocouples fixed in the surface or moved along energy release length. Coolant temperature as measured in all cells at the model bundle outlet and also at the model inlet and outlet in the headers. As the modeling coolant the eutectic alloy sodium-kalium (22% Na + 78% К) is used because it has the Prandtl number close to the numerical value of the lead Prandtl number. It ensured identity of heat exchange processes occurring at the contact fuel rod coolant interface in the case hen coolants considered ere clear and hen there ere not thermalchemical phenomena at the heat exchange surface. Thermal modeling of fuel rods of the BREST-OD-300 reactor (fissionable fuel is uranium or plutonium mononitride, the cover is stainless steel, the interlayer is lead) as rather strict (accuracy - 5%) for the fourth harmonics of temperature field Fourier series expansion (к 0 = 4) being the main harmonics for the regular square rod arrangement. 1 Copyright by JCME 1

2 Spacing of model rods as carried out by one or to spacer grids (Figure 2, a), located from a beginning of energy release on distance l p = 372 and 672 mm accordingly. The construction of the spacer grid is characterized by a frame hich is made of plates inserted one into another. The plates form square cells. The distance beteen spacer grids is 300 mm, it simulates a disposition of spacer grids in the core of the a) 3. EXPERIMENTAL RESULTS Regular model bundle The Nusselt numbers stabilized alone length for smooth pins are described by the formula [3, 4]: a) (1) Nu = 7,55 s/d - 14(s/d) -5 0,64+0,246 s/d + 0,007 Pe 1,20 s/d 1,50; 10 Pe For pins ith one spacer grid the formula for Nu numbers is similar to the formula (1), but it has other factor for Pe b) b) number [4, 5]: Nu = 7,55 s/d - 14(s/d) -5 + а Pe 0,64+0,246 s/d, (2) here а=0,01 for overlapping of passage cross-section for the coolant ε р = 10% и а = 0,009 for ε р = 20%. The important factor in temperature field along length of the model pin is the absence of of a all of the BREST reactor. Heated length of assembly as 960 mm. Figure 1. Construction (а) and cross-section (b) of model subassembly ith the regular lattice of model pins. 1 gasket obturating, 2 thermocouples outlet, 3, 10 upper and loer header, 4 lattice of thermocouples, 5, 8 upper and loer centering lattices, 6 model vessel, 7 model rods, 9 guiding vessel, 11 poer supplier, 12 poer supplier obturating, 13 square cover, 14 rotary (measuring) model fuel rod, 15 support bolt, 16 vessel; r 1 r 9 the radiuses, on hich the thermocouples equidistant from center of bundle are located Figure 2. Construction of the spacer grid (a); cross-section scheme of the model assembly ith to groups of model rods ith different diameters (d 1 > d 2 ) for different heat flos of groups of model rods (b) model pin under the spacer grid; moreover, in the region of the grid temperature decrease of the all is observed, and for ε р = 20 % happens more noticeable than for ε р = 10 % temperature loering of the all of the model pin in the region of the spacer grid (Figure 3). 2 Copyright 2003 by ASME

3 Figure 3. Comparison of temperature pressures in the l=420mm region of the spacer grid ( t ) in relation to the α stabilized values of temperature pressures ( t α stab ) for various Pe numbers in the bundles ith the spacer grids ε р = 20 ( ) and 10% ( ). The periodic non-uniformity of temperature along the perimeter of the pins corresponds to the cosine la (Figure 4, a, b). Values of dimensionless maximum non-uniformities of max min temperature T = ( t t ) λ / qr (here λ f thermal conductivity of coolant, q - average heat flux along pin perimeter, R - external radius of fuel pin) at decrease of the Peclet numbers tend to values of temperature non-uniformities for laminar mode of flo, characteristic for pitches (s/d) and parameters of equivalent thermal conductivity of fuel pins of the BREST-OD-300 reactor (Figure 5). Figure 4, a, b. Variation of non-uniformities of temperature along a perimeter of the measuring pin in the model bundle ith s/d = 1,25 in dimensional (а) and dimensionless (b) form for Pe = The generalized formula for various zones of fuel pins of the BREST-OD-300 reactor is recommended: Tл T = s / d Pe (3) 1.24 s/d 1.34; 1 Pe Values of T л are obtained from the calculated data. f a) b) Figure 5. Dependence of maximum non-uniformities of temperature of model pins from the Peclet number in model bundle ith s/d = 1,25 ( ), 1,28 ( ) and 1,34( ) (equivalent thermal conductivity ε 4 =1,4); -calculation using formula (3), T л values under the corresponding s/d for laminar mode of flo Heat transfer of fuel pins ith to spacer grids (ε р = 20 %) is a little bit higher (approximately 15 %) then heat transfer for fuel pin ith one spacer grid [6]. For an investigated range of Peclet numbers universal criteria dependence (2) is recommended in hich factor a for considered case is 0,0115. In the field of small Peclet numbers, there is a passage to the limit of Nusselt numbers to values of laminar mode of coolant flo in the subassembly ithout spacer grids. According to formula (2) groth of heat transfer in accordance ith transition from smooth pins to pins ith one and then ith to spacer grids is the result of, as already it as marked, the turbulent component in the Nusselt number, that is reflected by values of the factor a for one (a = 0,009) or to (a = 0,0115) spacer grids on comparison ith smooth pins (a = 0,007). For small flo rates of coolant the effect from availability of spacer grids (one or several) in the relation of heat transfer is absent practically. Irregular model bundle ith one spacer grid Main regularities for temperature fields of the measuring model pin located on the boundary of zones ith various diameters and poers of model pins (Figure 2, b) [7, 8] let us consider for an experiment ith small flo rate of coolant (Re = 3030) and large ratio of poer of zones (N 15 / N 10 = 2,0), hen the characteristic regularities are exhibited most precisely (here N 15 and N 10 - poer of model pins in zones of model assembly ith an amount of model pins 15 and 10 accordingly). Periodic non-uniformity of temperature. At small distances from a beginning of energy release (up to the spacer grid) coolant heating in cells round the measuring model pin differ eakly even for large distinction poers of model pins (N 15 / N 10 = 2,0) in zones ith s/d 1 =1,25 and s/d 2 =1,46. It creates conditions for appearance of periodic nonuniformities of temperature along the perimeter of the model pin. The dimensionless non-uniformities of temperatures 3 Copyright 2003 by ASME

4 reduced to a cell are changed under the cosine la (similarly to Figure 4, b) and do not reveal any certain dependence on value of jump of energy release. The dimensional non-uniformities depend on jump of energy release at zones as the modification of relative energy release N 15 / N 10 assumes modification of value of specific heat flux at the surface of adjacent model pins. General non-uniformity of temperature. In an item of the thermocouple 3 (and further along zone of energy release) the general non-uniformity of temperature along the perimeter of model pin stipulated by a difference of heating of coolant at «boundary» cells for distinguished poers of pins in zones is exhibited. The general non-uniformity of temperature along the perimeter of the measuring model pin is determined by temperature difference in points of the perimeter ϕ=0 and (maximum - at ϕ=0 0, minimum at ϕ=180 0 ) (Figure 6, a). And on the contrary, if the poer in the zone ith c s/d 2 =1,46 considerably exceeds poer in the zone ith s/d 1 =1,25 (N 10 / N 15 =2 reverse jump of energy release to considered jump), the general non-uniformity of temperature is determined by maximum at ϕ=180 and minimum at ϕ=0 0. In some operational modes of model assembly (especially for transitional current of coolant from laminar to turbulent) effect of displacement of temperature maximum along the perimeter of the model pin in different cross-sections along length of the zone of energy release (Figure 6, b) *) occurs: the thermocouple 8 fixes a profile ith maximum at ϕ=0 0, thermocouple 9 at ϕ=60 0 ; 10 at ϕ=120 0 etc. (increase of number of the thermocouple corresponds to increase of distance from a beginning of energy release). Maximum general non-uniformities of the measuring model pin are illustrated in Figure 7 (the dimensionless form) as the function of jump of energy release at fixed flo rate (Pe, Re numbers) of coolant (the field of temperature is shon in Figure 6, a). As it is visible, these are linear dependences (more strict at greater flo rates of coolant), demonstrating increase of non-uniformities ith increase of the relative jump of poer N 15 / N 10 and ith decrease of Pe (Re) number. At reverse jumps of poer hen the redistribution of position of a maximum and minimum of temperature at the points ϕ=0 и happens, the non-uniformity of temperature is conditionally postponed donards from zero on the ordinate axis and the sign «minus» is given to it. The modification of relative energy release in adjacent zones of model pins of model subassembly is investigated in a more broad interval than it can be in the core of the reactor BREST- OD-300; the experimental data for Pe numbers of the reactor (~ 2000) are reproduced by a reliable extrapolation; the final recommendations for the reactor base on use of the greatest values of general non-uniformities of temperature from the indications of thermocouples located at the outlet of the zone of energy release of the model subassembly. *) Pe and Re numbers used in Figure 6, b, and further (Figure 7, 8) are calculated from average velocity in the cross-section of assembly and from hydraulic diameter of a regular cell in the zone ith s/d 1 = 1,25. The velocity in the cell s/d 1 = 1,25 is practically equal to the velocity to velocity. Figure 6. Comparison of temperature fields around the perimeter of the measuring model pin ith one and to spacer grids (a); variation of surface temperature of the measuring model pin around the perimeter in the various cuts along length of the zone of energy release (thermocouple 8-12) for Pe =63 and N 15 / N 10 =1,0 (b), R - external radius of the model pin, q average heat flux around the perimeter of the model pin In the core of the reactor BREST-OD-300 the greatest nonuniformities of temperature also ill take place at the outlet of zone of energy release here distinction of heating in adjacent cells is maximum. Irregular model bundle ith to spacer grid Interaction of adjacent spacer grids can reduce temperature non-uniformities in the fuel subassembly (especially in irregular zones), increase heat transfer (it as noted above) and as a hole to have a positive effect on temperature modes of fuel pins of the reactor BREST-OD-300. Characteristics of a temperature field around the perimeter of the model pin (Figure 6, а) is (as ell as in the experiments ith one spacer grid) maximum of surface temperature of the model pin in the narro zone of model assembly (s/d = 1,25, narro gap beteen model pins - - ϕ=0 0 ) and minimum in the ide zone of model assembly (s/d = 1,34, gap beteen model 4 Copyright 2003 by ASME

5 pins at ϕ=180 0 )*). It determines general non-uniformity of temperature along the perimeter of the model pin stipulated by of the coolant in the zone ith s/d = 1,25 and underheating in the zone ith s/d = 1,34. T 1 t = max t qr min λ f In Figure 6, a the average temperature fields in the second half of model subassembly from the indications of several thermocouples are illustrated hen the distributions are characterized by smaller value of temperature non-uniformity on comparison ith non-uniformity at the outlet of the zone of energy release. The experiments have shon that a ratio beteen non-uniformities of temperature of compared variants (subassembly ith one and ith to spacer grids) is the same for various cross-sections of the bundle (including for the crosssection at the outlet of the zone of energy release representing the greatest practical interest). This ratio remains the same for illustrated in Figure 6, a average values of non-uniformities of temperature from indications of several thermocouples. Determined from the experiments ith one spacer grid the linear dependence of non-uniformity of temperature on relative poer N 15 /N 10 in a broad band of a modification N 15 /N 10 (0,5 2,0) in general is confirmed by experiments ith to spacer grids, though the range of modification of relative poers in these experiments as essentially less (0,82 N 15 /N 10 1,20). It allos to conduct recalculation of obtained regularities for relative poer N 15 /N 10 = 1,0 to other values N 15 /N 10. The dependence of dimensionless non-uniformity of temperature on Peclet number for using both one spacer grid ( T 1 ) and to spacer grids ( T 2 ) is shon in Figure 8. It is visible, that the non-uniformities are not so hardly changed in the field of large Peclet numbers (Pe > 700), but sharply increase ith decrease of Peclet number in the area Pe< (rate of increase T 1 and T 2 is about identical). The non-uniformity T 2 is 0,52 T 1 for Pe 100 and 0,42 T 1 for Pe Dependence T 2 / T 1 on Pe is approximately linear. It is possible to accept approximately T 2 0,47 T 1 in investigated range of modification of Peclet numbers (100 Pe 1300). Figure 7. Dependence of general non-uniformities of temperature around the perimeter of the measuring model pin on jumps of energy release N 15 / N 10 and Peclet numbers (experiment ith one spacer grid) *) The intermediate cell beteen zones s/d1 = 1,25 and s/d 2 =1,46 is considered as conditionally regular cell if its relative pitch s/d = 1,34 is calculated using average diameter d = (d 1 +d 2 )/2. Figure 8. Dependence of dimensionless temperature nonuniformity on Peclet number for using one ( T 1 ) and to ( T 2 ) spacer grids For analysis of the case under consideration for the core of the BREST-OD-300 reactor it is necessary to use the values of T 1 from the nomogram represented in Figure COMPUTATIONAL RESEARCH Calculation research of thermohydraulic characteristics for the core of the BREST-OD-300 is carry out using computer codes BRS-AZ, BRS-TVS.R and TIGR-BRS [4, 9, 10]. The codes are based on the subchannel technique and take into account geometric characteristics of fuel subassemblies of the cores of the BREST-type reactors (square arrangement of fuel pins, presence of irregular cells formed by displacers, supporting tubes etc.) and property of lead as coolant. In the frame of the subchannel technique fuel subassembly is presented as the system of connected cells for each of hich the 5 Copyright 2003 by ASME

6 equations of balance of mass, impulse and energy are solved ith regard to thermal and hydrodynamic interaction at the boundaries of the cells. The equation are solved for the case of the stabilized mode of coolant flo in the code TIGR-BRS. Initial conditions for the given set of equations are the conditions at the inlet of a fuel subassembly (temperature of the coolant and distribution of velocity at the inlet (or average velocity of the coolant in the fuel subassembly)). In the codes BRS-AZ and BRS-TVS.R non-stabilized mode of flo is considered. For closing of the set of equations in the finite difference form the system of closing relations is developed including relations for coefficients of hydraulic resistance, coefficients of interchannel exchange, coefficients of heat exchange. In the frame of the designed coded the unit is created for determination of temperature distribution in fuel pin in hich non-stationary unaxisymmetrical heat conduction equation in the fuel pin is solved. The equation in cylindrical coordinates is solved by the alternating direction method. Verification of the codes is carried out using described above experimental results obtained for model subassemblies of the BREST-OD-300. The comparison of computational and experimental results is carried out for cases hen the model subassembly consists of model pins ith identical diameters and ith identical energy release and hen the model subassembly consists of to zones of model pins ith identical diameter but ith different energy release. The good agreement of computational and experimental results is obtained (not orse ~ 5 % from the heatings of the coolant) both for uniform energy releases under the cross-section of model subassembly (Figure 9, а) and for jumps of energy release (Figure 9, b). The substantiation of permissible limit of thermohydraulic parameters of the core at stationary modes is conducted by to stages. At the first stage ith the help of programs BRS-AZ calculation of the core and the reflector is executed. At this stage most heat-stressed fuel subassemblies are determined for further calculation and also distribution of the exchange coefficients in the equations of conservation for these fuel subassemblies. At the second stage using the code BRS-TVS.P thermohydraulic calculation of most heat-stressed fuel subassembly are conducted. The cross-section of the subassembly is partitioned to cells formed by fuel pins and nonheated tubes of the subassembly. At this stage obtained information is: about distribution of average values of velocity and temperature of the coolant in the cells; distribution of temperature in the fuel pins ith regard to and ithout regard to factors of. The computational results of thermohydraulic characteristics of the BREST-OD-300 reactor core obtained on the second stage are indicated in the table. As a result of thermohydraulic computational research it is shon that three-zonal profiling of fuel loading at using of zones of fuel pins ith different diameters but ith the same pitch beteen fuel pins allos to align maximum values of temperatures of fuel pin clads in the conditions of the core ithout shrouds. The most critical parameters of the core cooled by lead coolant are the maximum temperature of external surface of fuel pin clad and its azimuth nonuniformity. Table. Thermohydraulic characteristics of the core for zones of fuel loading for rated poer Zones of fuel loading Performance Central Inter- mediate Peri- pheral Maximum linear poer of fuel pins, W/sm Maximum density of 1,54 1,37 1,01 heat flux, МW/m 2 Heat transfer, kw/(m 2 K) Reynolds number 1, , , Peclet number Maximum velocity of the coolant, m/s 1,79 1,72 1,54 of the coolant at the outlet of the core, С : of external surface of fuel pin clad, С: of internal surface of fuel pin clad, С: of lead in the gap, С: of fuel, С: Maximum azimuth nonuniformity of temperature of clad, С: The problems of further thermohydraulic research are to study various kind of thermohydraulic heterogeneities in the subassemblies caused by availability of fuel pins ith different diameters and energy release, supporting tubes and displacers, contact of the core ith the side reflector, temperature fields of the deformed lattice of fuel pins, factors of, 6 Copyright 2003 by ASME

7 influence of contact thermal resistance to temperature modes of fuel pins etc. a) the model fuel assembly for the lead cooled fast reactor, Proc. Scientific Session MIFI-2000, MIFI, Mosco, Vol. 8, pp (in Russian) [4] Kuzina, J.A., 2001, Thermohydraulic problems of reactors ith inherent safety features, Proc. XIII School-Seminar of Yong Scientists and Specialists under the leadership of the Academician, Professor A.I. Leontiev Physical Principles of Experimental and Mathematical Simulation of Heat and Mass Transfer and Gas Dynamics in Poer Plants, S.-Petersburg. Vol. 2, pp (in Russian) [5] Zhukov, A.V., Kuzina, J.A., Sorokin, A.P. etc., 2002, Experimental Research of Heat Transfer in the Core of Lead Cooled BREST-OD-300 Reactor Using Models, Thermal Engineering, 3, pp (in Russian) b) (ros of the cells in the model bundle) Figure 9. Temperature field along length of the model pin (a) ( - computational data, experimental data, t α р = С, t α э = С) and distribution of average heating in the ros of the cells in the model subassembly ith s/d = 1,46 for jump of energy release N(15)/N(10) = 2 (b) ( - experiment, - calculation) REFERENCES [1] Orlov, V.V., 1991 Ne Stage of Nuclear Poer and Fast Reactors Cooled by Lead, Information Nesletter 3 (10). Nuclear society USSR, p. 6. (in Russian) [2] Adamov, E.O., Orlov, V.V., 1998 Development of nuclear poer on the basis of the ne concepts of nuclear reactors and fuel cycle, The program and abstracts of conference "Heavy liquid-metal coolant in nuclear technologies", Obninsk, IPPE, p.15. (in Russian) [3] Kuzina, J.A., Smirnov, V.P., Zhukov, A.V., and Sorokin, A.P., 2000, Research of temperature fields and heat transfer on [6] Kuzina, J.A., Sila-Novitsky, A.G., 2002, Model Experiments and Calculations (TIGR-BRS code) for Study of Temperature and Velocity Fields in the Cores of the Reactors ith Heavy Coolant, Thermal Engineering, 11. (in Russian) [7] Kuzina, J.A., Zhukov, A.V., and Orehov, M.V., 2001, Research of Temperature Fields for Fuel Rods Located on the Interface of Subzones ith Different Diameters and Energy Release of Fuel Rods (for Reactors such as BREST), Abstracts VII Int. Conf. NPP Safety and Personnel Training, Obninsk, pp (in Russian) [8] Kuzina, J.A., Zhukov, A.V., Orehov, M.V. etc., 2002, Temperature Fields of Fuel Pins in The Core of the BREST- OD-300 Reactor (Experiments Using Model Assemblies) Proc. Scientific Session MIFI-2002, MIFI, Mosco, Vol. 8, pp (in Russian) [9] Kuzina, J.A., Smirnov, V.P., and Sorokin, A.P., 2002, Computational Research For Thermohydraulic Substantiation of the Core for the BREST-OD-300 Reactor, Proc. Scientific Session MIFI-2002, MIFI, Mosco, Vol. 8, pp (in Russian) [10] Smirnov, V.P., Kuzina, J.A., Papandin, M.V., etc., 2002, Complex of Program Codes of Thermohydraulic Calculation of the Core for the BREST-OD-300 Reactor Abstract of conference Thermal Physics Heat and Mass Transfer and Properties of Liquid Metals, Obninsk, pp (in Russian) 7 Copyright 2003 by ASME