Nội dung

Nuclear Science and Technology, Vol. 4, No. 1 (2014), pp. 10-25 Design Analyses for Full Core Conversion of The Dalat Nuclear Research Reactor Luong Ba Vien, Le Vinh Vinh, Huynh Ton Nghiem, Nguyen Kien Cuong Reactor Center – Nuclear Research Institute – Vietnam Atomic Energy Institute 01 Nguyen Tu Luc, Dalat, Lamdong Email: reactor@hcm.vnn.vn (Received 5 March 2014, accepted 10 March 2014) Abstract: The paper presents calculated results of neutronics, steady state thermal hydraulics and transient/accidents analyses for full core conversion from High Enriched Uranium (HEU) to Low Enriched Uranium (LEU) of the Dalat Nuclear Research Reactor (DNRR). In this work, the characteristics of working core using 92 LEU fuel assemblies and 12 beryllium rods were investigated by using many computer codes including MCNP, REBUS, VARI3D for neutronics, PLTEMP3.8 for steady state thermal hydraulics, RELAP/MOD3.2 for transient analyses and ORIGEN, MACCS2 for maximum hypothetical accident (MHA). Moreover, in neutronics calculation, neutron flux, power distribution, peaking factor, burn up distribution, feedback reactivity coefficients and kinetics parameters of the working core were calculated. In addition, cladding temperature, coolant temperature and ONB margin were estimated in steady state thermal hydraulics investigation. The working core was also analyzed under initiating events of uncontrolled withdrawal of a control rod, cooling pump failure, earthquake and MHA. Obtained results show that DNRR loaded with LEU fuel has all safety features as HEU and mixed HEU-LEU fuel cores and meets requirements in utilization as well. Keywords: HEU, LEU, neutronics, thermal hydraulics, safety analyses I. INTRODUCTION In this full core conversion study, neutronics, thermal hydraulics and safety analysis were carried out to investigate characteristics of LEU working core fully loaded with LEU fuel. All computer codes were validated with HEU and mixed cores. Using MCNP [6], REBUS-PC [5] and VARI3D computer codes, a series of static reactor physics calculation were performed to obtain neutronics parameters of the working core (see Fig. 1). Some parameters included in the design of working core with shutdown margin, excess reactivity taking into account of irradiated Beryllium poisoning, control rod worths, detailed power peaking factors, neutron performance at the irradiation positions, reactivity feedback coefficients, and kinetics parameters. Because the higher content of 235U in a LEU FA compared to HEU FA, it is needed to rearrange the fuel assemblies and berrylium rods with the different way to the first HEU core to meet the safety requirements. Thermal hydraulics parameters at steady state condition were obtained by using PLTEMP3.8 code [11] introduced models and correlations that suitable for the concentric tube fuel type and natural convection regime of the DNRR. Based on the neutronics analysis parameters of the LEU core, the postulated transients and accidents selected for the DNRR are analyzed. The RELAP5/MOD3.2 code [15] was used for analysis of RIA (Reactivity Initiated Accident), LOFA (Loss Of Flow Accident) transients. ©2014 Vietnam Atomic Energy Society and Vietnam Atomic Energy Institute LUONG BA VIEN, LE VINH VINH, HUYNH TON NGHIEM, NGUYEN KIEN CUONG These study results showed that a LEU core loaded with 92 fuel assemblies and 12 beryllium rods around the neutron trap satisfies the safety requirements while maintaining the utilization possibility similar to that of the previous HEU and recent mixed fuel cores. REBUS-MCNP Linkage [7] was used to calculate burnup distribution using “two way” linking option in which MCNP is used for calculating neutron flux and cross section in one group neutron energy and burn up calculation is implemented by REBUS-PC. The MCNP5 code using an ENDF-B/VI cross section library was used to construct a detailed geometrical model of each reactor component and calculate control rod worths, multiplication coefficient, power distribution, neutron flux performance in irradiation positions, reactivity feedback coefficients, and kinetics parameters (prompt neutron life time and delayed neutron fraction). A detailed geometrical model of reactor components including all fuel assemblies, control rods, irradiation positions, beryllium and graphite reflectors, horizontal beam tubes and thermal column was made in the MCNP model, except in the axial reflectors above and below the fuel assembly where some materials were homogenized. Fig. 2a provides the radial and axial models of the reactor for Monte Carlo Calculations. Fig. 1. The new designed working core loaded with 92 LEU FA and 12 Beryllium rods. II. CALCULATION MODELS AND COMPUTER CODES A. Neutronics Calculation and Thermal Hydraulics The diffusion code REBUS-PC with finite difference flux solution method was used to perform core calculation for reactor physics characteristics and operation cycle calculations with micro neutron cross sections according to 7 energy groups (collapsed from 69 energy groups) that were generated by WIMS-ANL code [4]. The FA cross sections were generated in a radial geometry with each fuel element depleted based upon its unique neutron spectrum in the WIMS-ANL model. The REBUS-PC fuel depletion chains included production of six Pu isotopes, Am-241, Np237, and lumped fission product. Isotopic precursors of Xe-135 and Sm-149 were also included in the depletion chains so that Xe and Sm transients during periods of shutdown and startup could be modelled. The kinetics parameters were calculated also by VARI3D code. The real and adjoint fluxes which are required to compute these parameters were provided by DIF3D-a main module of REBUS-PC code. In diffusion theory, the reactor was modeled in hexagonal geometry with a heterogeneous representation of the fuelled and non-fueled portions (see Fig. 2a). Each homogenized fuel assembly was modelled using five equal volume axial depletion zones. The beam tubes were modeled using a homogenized mixture of air or concrete, graphite and aluminum. The reactor models for diffusion and Monte-Carlos computer codes were validated by comparing with good agreement not only to 11 DESIGN ANALYSES FOR FULL CORE CONVERSION OF … the fresh HEU configuration cores but also to the HEU burnt cores. These models were then applied for partial core conversion analyses of DNRR [3]. The measured data collected during the deployment of partial core conversion project showed that the predicted calculation results are quite acceptable [8,9]. for PLTEMP code. A fuel assembly was modelled as three concentric cylindrical tubes. Before using PLTEMP code to calculate for DNRR with fully LEU fuel assemblies, the code was validated by comparing analytical results with experimental results of mixed-core. The PLTEMP/ANL3.8 [15] thermalhydraulics code for plate and concentric-tube geometries with capability of calculating natural circulation flow was used for thermalhydraulics analyses. A chimney model as well as Collier heat transfer correlation and CHF Shah’s correlation have been recently implemented make the code suitable DNRR thermal-hydraulics calculation. B. Transient/Accidents analyses The DNRR has three barriers as other research reactors that prevent or limit the transport of fission products to the environment, which are fuels and cladding, reactor pool water and reactor confinement. The safety system settings are showed in Table I. Fig. 2b shows the model of WWR-M2 fuel assembly, core and chimney of the DNRR Table I. Safety system settings. Parameters Maximum thermal power (Pmax) Minimum reactor period (Tmin) Deficient level of pool water Primary coolant flow rate Secondary coolant flow rate Safety system settings 550 kW (110% FP) 20s 60 cm 40 m3/h 70 m3/h In the Safety Analysis Report (SAR) for the DNRR [1], the possible initiating events were classified by groups. The initiating events in each group are then analyzed and justified in order to identify the limiting event that will be selected for further detail quantitative analysis. The limiting event in each group has potential consequences that exceed all others in that group. Limiting events were selected for detailed analyzed are as follows: (1) Uncontrolled withdrawal of a control rod; (2) Primary/Secondary Pump Failure; (3) Earthquake; (4) Fuel cladding failure. A summary of the core parameters used for the safety analysis is given in Table II. Fig. 2a. Radial and Axial models for Monte Carlo calculations (upper) and Radial model for Diffusion Theory calculations (under). 12 LUONG BA VIEN, LE VINH VINH, HUYNH TON NGHIEM, NGUYEN KIEN CUONG To ensure the fuel clad integrity in operational condition and to protect the public and the environment in case of accident, in the SAR for the DNRR, the following acceptance criteria were defined: For occurrences: anticipated 1) operational 3) (1) Minimum margin to departure from nucleate boiling (DNB) shall be over 1.5; (2) Maximum temperature of fuel cladding shall not exceed 400oC; (3) Fuel assured. cladding integrity shall be - For accident conditions: (1) Core covering shall be maintained; (2) Core damaged; shall not be remarkably (3) Release of fission products into the environment shall not be remarkable. The RELAP5 code was used for analyzing the events of excess reactivity insertion by uncontrolled withdrawal of a control rod and earthquake. The piping of the primary cooling system and pool volume were divided into nodes with similar dynamic characteristics. The reactor core was divided into 2 channels with axial nodes. The hot channel represents the hottest channel in the 2) Fig. 2b. DNRR model for PLTEMP (1-fuel assembly cross-section; 2-FA model for PLTEMP; 3-reactor coolant system model). core corresponding to a cooling channel with maximum heat flux. The average channel represents the rest of the cooling channels. Each channel was modelled as three fuel element plates and four coolant flow gaps. The nodding diagram of the DNRR for RELAP5/3.2 is presented in Fig. 2c. The MACCS2 code [19] was used to estimate the radiological impact of the hypothetical accident on the surrounding public. The core radiation inventories were calculated by ORIGEN2 code [20] using neutron cross-sections of the actinides obtained from MCNP5 code. Fig. 2c. Nodding diagram of DNRR for RELAP5/3.2. 13 11 DESIGN ANALYSES FOR FULL CORE CONVERSION OF … Table II. Core parameters used for safety analysis. Parameters Power, kW Coolant inlet temperature, oC Peaking factor (shim rods at 300 mm) - Axial peaking factor - Radial peaking factor - Local peaking factor Reactor kinetics - Prompt neutron life, s - Delayed neutron fraction (1$) Temperature reactivity coefficients - Moderator, %/K; (293-400oK) - Fuel, %/oC; (293-400oK) (400-500oK) (500-600oK) - Void, %/% of void (0-5%) (5-10%) (10-20%) Reactivity control - Shutdown worth, % (2 safety rods) - Maximum withdrawal speed of one shim rod, mm/s and of the regulating rod, mm/s Reactor protection characteristics - Response time to overpower scram, s - Response time to fast period scram, s Start-up range Working range - Drop time of control rods, s Values 500 32 1.363 1.376 1.411 8.92510-5 7.55110-3 - 1.26410-2 - 1.8610-3 - 1.9210-3 - 1.5610-3 -0.2432 -0.2731 -0.3097 3.7 3.4 20 0.16 9.1 6.7 0.67 results in large negative reactivities which alter flux and power distributions. III. RESULTS AND DISCUSSIONS A. Neutronics and Thermal Hydraulics Program Beryl [10] has been modified to calculate the 3He, 6Li and 3H concentrations. The MCNP5 was then used to determine the poisoning effect of 3He, 6Li and 3H concentrations on reactor core reactivity. The comparison of reactivities between calculation results and measured data of some beryllium blocks irradiated in DNRR (Table III) shows that the negative reactivity of irradiated beryllium determined by above-mentioned 3 He and 6Li Poisoning of Irradiated Beryllium [10] Since 1984, the DNRR has been put into operation with a considerable amount of Beryllium used for neutron trap at the core center and periphery for improving neutron reflection around. Because Beryllium has large thermal neutron absorption cross sections, the buildup of 3He, 6Li and 3H concentrations 14 LUONG BA VIEN, LE VINH VINH, HUYNH TON NGHIEM, NGUYEN KIEN CUONG method is reliable. Six beryllium rods were used for measurement, two fresh beryllium rods and four irradiated beryllium rods (two beryllium rods at the end 1994 and two at the end 2002). 9-6 and 5-6 positions were chosen to measure reactivity of couple beryllium rods through changing position of control rod (Regulating Rod). The error of control rod position is estimated about 0.4 cent. Following calculation scheme for beryllium poisoning above, reactivity of the poisoning process in new configuration cores about -1$. All calculation for design LEU cores, beryllium poisoning is included in the model for MCNP code. Table III. Comparison of calculated and measured of reactivities of irradiated beryllium rods in DNRR. Measured reactivity (Cent) Calculated reactivity (Cent) Error (%) 2 Beryllium Rods at the end 1994 -3.89  0.4 -4.65  0.0038 16.34 2 Beryllium Rods at the end 2002 -6.28  0.4 -7.19  0.0039 12.66 The working core characteristics From the calculation results of shutdown margins, excess reactivities, power peaking factors, and neutron performance at the irradiation positions of 4 candidates cores, the working core with the better features from the safety and utilization point of view was chosen for detailed analysis. The main calculated characteristics of working core is showed in the Table IV. The shutdown margins of the core is met the safety requirement of -1.0%. Calculated neutron flux at the neutron trap of the core is nearly the same as that of mixed core (92HEU+12LEU). Table V shows the control rod worths. Detailed neutron flux performance at the main irradiation positions are presented in Table VI. Table IV. Calculation results of working core compared with current mixed core. Parameters LEU Core Excess Reactivity (%) – Fresh Excess Reactivity (%) – After 600FPDs Shutdown Margin (%) – Fresh Shutdown Margin (%) – After 600 FPDs Radial Power Peaking Factor Control Rods Out Control Rods In Thermal Neutron Flux at Neutron Trap Center (n/cm2) Control Rods Out Control Rods In Fast Neutron Flux at Neutron Trap Center (n/cm2) Control Rods Out Control Rods In 6.63 3.79 -2.92 -6.62 15 11 1.398 1.434 Current Mixed Core -4.56 1.431 2.22E+13 2.14E+13 2.22E+13 1.95E+12 1.92E+12 3.15E+12 DESIGN ANALYSES FOR FULL CORE CONVERSION OF … Table V. Control Rods worths (%k/k). Control Rods Shim rod 1 Shim rod 2 Shim rod 3 Shim rod 4 Regulating rod Safety rod 1 Safety rod 2 Core1 Fresh 2.5896 2.6100 2.7784 2.4687 0.4363 2.1955 2.2356 MCNP error 0.000091 0.000111 0.000118 0.000122 0.000126 0.000106 0.000119 Core1 Burnt 2.3539 2.4033 2.5381 2.2604 0.3629 2.3084 2.3579 MCNP error 0.000091 0.000124 0.000122 0.000117 0.000119 0.000115 0.000105 Table VI. Neutron flux performance. Maximum Average Maximum Average Maximum Average Maximum Average Fresh 2.07E+13 1.45E+13 9.45E+12 7.00E+12 5.41E+12 4.11E+12 9.24E+12 6.85E+12 Burnt 2.20E+13 1.49E+13 9.86E+12 7.12E+12 5.66E+12 4.18E+12 9.71E+12 7.01E+12 Epithermal, <0.821MeV (n/cm2.s) Fresh Burnt 6.79E+12 7.12E+12 6.00E+12 6.04E+12 8.19E+12 8.42E+12 6.53E+12 6.51E+12 9.63E+12 9.76E+12 7.23E+12 7.15E+12 8.02E+12 8.22E+12 6.41E+12 6.40E+12 Average 3.55E+12 3.56E+12 7.58E+11 7.56E+11 Irradiation positions Neutron Trap Channel 13-2 Channel 7-1 Channel 1-4 Rotary Specimen Thermal, <0.625eV (n/cm2.s) Power Distribution and Power Peaking Factors Fast, <10MeV (n/cm2.s) Fresh 1.83E+12 1.62E+12 2.98E+12 2.46E+12 4.22E+12 3.19E+12 2.92E+12 2.42E+12 Burnt 1.92E+12 1.63E+12 3.02E+12 2.44E+12 4.26E+12 3.15E+12 2.99E+12 2.40E+12 1.93E+11 1.93E+11 of control rods at 250 mm. Detailed axial power distribution according to control rod position was also calculated. Radial power distributions at different control rod position are showed in Fig. 3. Power peaking factors of the core with different position of control rods were calculated and presented in Table VII. The maximum power peaking factor is in position Table VII. Power peaking factor according to control rod positions Position (mm) 0 150 200 250 300 350 600 F.A. Radial 1.378 1.378 1.375 1.377 1.376 1.378 1.378 Peaking Factor Core Radial Axial 1.398 1.296 1.399 1.343 1.403 1.356 1.409 1.365 1.411 1.363 1.415 1.336 1.434 1.284 16 Total 2.498 2.589 2.615 2.648 2.646 2.605 2.537 LUONG BA VIEN, LE VINH VINH, HUYNH TON NGHIEM, NGUYEN KIEN CUONG 0.973 0.947 0.913 0.901 0.877 0.875 0.917 0.906 0.962 0.939 1.138 1.090 0.998 0.974 1.018 0.991 1.020 0.992 0.979 0.964 1.004 1.008 1.090 1.126 1.164 1.129 1.165 1.126 1.107 1.085 SR 1.410 1.370 1.006 1.001 1.021 1.023 1.406 1.082 1.368 1.124 1.281 1.296 1.259 1.283 ShR 0.810 0.863 1.016 0.994 1.106 1.145 0.860 0.910 0.918 0.917 1.421 1.408 1.381 1.368 1.038 1.038 1.031 1.019 1.198 1.156 SR 1.124 1.097 1.005 1.006 0.986 0.970 1.180 1.137 1.079 1.114 0.816 0.866 0.858 0.857 0.985 0.960 ShR 0.745 0.803 0.843 0.846 0.959 0.933 RgR 0.843 0.837 0.808 0.818 0.755 0.808 0.919 0.903 0.865 0.855 0.911 0.900 0.918 0.919 0.868 0.919 1.252 1.273 0.929 1.296 0.975 1.312 0.921 0.915 0.842 0.850 0.930 0.929 1.139 1.122 1.307 1.284 0.908 0.913 0.787 0.843 0.882 0.937 1.313 1.289 1.353 1.321 0.963 0.958 0.775 0.833 ShR 1.220 1.192 0.830 0.881 0.968 0.959 0.885 0.895 0.835 0.889 0.996 1.364 0.987 1.332 0.903 0.904 0.996 0.977 0.858 0.858 0.825 0.876 ShR 0.906 0.953 0.983 0.975 1.056 1.027 1.167 1.117 0.762 0.817 0.790 0.841 0.872 0.873 0.849 0.851 0.980 0.957 0.911 0.887 0.845 0.835 0.841 0.836 0.831 0.836 0.895 0.889 0.973 0.949 Fig. 3. Radial power distribution (Upper values: Fresh Core; Under values: Burnt Core) Reactivity Feedback Coefficients and Kinetics Parameters kinetics parameters of the LEU cores calculated using the VARI3D and MCNP5 codes. The calculated results from the two computer codes are in good agreement. These data will be used in transient calculation for safety analysis of fully LEU core of DNRR. Reactivity feedback coefficients calculated with the MCNP5 are depicted in Table VIII. The negative results of reactivity feedback coefficients show the inherent safety of the LEU core. Table IX shows the Table VIII. Feedback reactivity coefficients. Parameter DATA ±σ -0.01317 0.00005 -0.00192 -0.00182 0.00005 0.00003 -0.00154 0.00002 -0.2514 -0.2784 -0.3255 0.0011 0.0012 0.0006 o Moderator Temperature Reactivity Coefficient (%/ C) 293 oK to 400 oK Fuel Temperature (Doppler) Reactivity Coefficient (%/oC) 293 oK to 400 oK 400 oK to 500 oK 500 oK to 600 oK Moderator Density (Void) Reactivity Coefficient (%/% of void) 0 to 5 % 5% to 10 % 10 % to 20 % 17 DESIGN ANALYSES FOR FULL CORE CONVERSION OF … Table IX. Calculated results of kinetics parameters for LEU core. Family, i Decay Const. λi (s-1) Relative Yield ai 1 2 3 4 5 6 1.334E-02 3.507E-02 3.273E-02 1.804E-01 1.208E-01 1.742E-01 3.030E-01 3.843E-01 8.503E-01 1.594E-01 2.856E+00 6.666E-02 Total delayed neutron fraction, β VARI3D MCNP5 – Fresh MCNP5 – Burnt Prompt neutron life time, ℓ Burn up calculation Fraction βi 2.648E-04 1.363E-03 1.315E-03 2.902E-03 1.204E-03 5.033E-04 7.551E-03 7.761E-03 7.762E-03 8.925E-05 extended about 11 years (calculated with 1300 hours per year) or 600 full power days (FPDs). The burn up of U-235 reached average value of 8.2% and maximum value of 11.4%. In the next cycle, about 8 fuel assemblies will be inserted so the reactor core will operate with 100 fuel assemblies. The Fig. 4 shows burn up distribution after 600 FPD operation. The first cycle length was estimated by REBUS-MCNP Linkage system code. Burn up calculations were performed by assuming that shim rods and regulating rod were in critical position following each burn-up step. The value of reactivity for Xe-135 poisoning was estimated about 1.2% k/k. The result of depletion shows that operating time may be Fig. 4. Burn up distribution using REBUS-MCNP Linkage system after 600 FPD. 18 LUONG BA VIEN, LE VINH VINH, HUYNH TON NGHIEM, NGUYEN KIEN CUONG working core meets the requirements of thermal hydraulics safety. At the power of 500kW with systematic errors, maximum cladding temperatures are below the permissible value of 103oC [2] and far below the ONB temperature (estimated about 116oC using Forster-Greif correlation). The maximum outlet coolant temperature is calculated about 60oC, much lower than saturated temperature (108oC). The PLTEMP code was used for calculating cladding temperature, coolant temperature and safety margins for the candidate cores. The calculated results are presented in Table X and Fig. 5. At nominal power without uncertainties and maximum permissible inlet temperature (32oC), the maximum cladding temperature is 90.50oC. Calculation was carried out for nominal power with systematic errors (equivalent to 70kW power) and the maximum cladding temperature is 95.69oC. In this case, by using Shah’s correlation, the obtained minimum DNBR is 9.9. The minimum flow instability power ratio (MFIPR) is 2.04. From above-mentioned calculated results, it may conclude that the Fig. 6 shows the comparison of cladding temperature of 92FA LEU cores and 89FA fresh HEU core. Compared to the 89FA fresh HEU core established in 1984, cladding temperature of working core is about 2oC lower. Table 10. Cladding temperature and ONB margin by PLTEMP Code. Distance (cm) 100.0 550kW with sys. error TTc(oC) ONB(oC) 68.95 47.39 76.58 40.14 85.63 31.51 92.89 24.48 97.33 20.06 99.23 17.97 98.43 18.41 94.41 21.94 89.94 25.90 84.43 30.74 78.79 35.57 74.64 38.98 600kW with sys. error TTc(oC) ONB(oC) 70.96 45.59 78.97 37.97 88.46 28.91 96.05 21.57 100.68 16.95 102.65 14.80 100.76 16.40 96.22 20.48 91.57 24.60 85.89 29.59 80.13 34.52 75.94 37.93 100 Temperature ( o C) Temperature ( o C) 2.5 7.5 12.5 17.5 22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 500kW without sys. error with sys. error TTTc(oC) ONB(oC) Tc(oC) ONB(oC) 63.91 51.89 66.89 49.24 70.56 45.59 74.13 42.36 78.46 38.07 82.71 34.18 84.83 31.90 89.61 27.50 88.77 27.95 93.85 23.26 90.50 26.05 95.69 21.25 89.86 26.34 94.95 21.63 87.10 28.58 91.91 24.13 83.98 31.14 88.24 27.24 79.67 34.76 82.92 31.91 74.91 38.73 77.42 36.64 71.21 41.70 73.32 40.02 90.0 80.0 T-clad 70.0 60.0 92 LEU FA Core2 95 89 HEU FA Core 90 92 LEU FA Core1 85 50.0 80 40.0 T-coolant 75 30.0 70 20.0 DT-ONB 65 10.0 0.0 0 10 20 30 40 50 60 60 0 Distance from core bottom (cm) Fig. 5. T/H parameters at 500kW without uncertainties. 10 20 30 40 50 60 Distance from core bottom (cm) Fig. 6. Comparison of calculated cladding temperature between 92FA LEU cores and HEU core. 19