This article was submitted to Nuclear Energy, a section of the journal Frontiers in Energy Research
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The 10 MW High Temperature Gas-cooled Reactor-Test Module (HTR-10) is the first High Temperature Gas-cooled Reactor (HTGR) in China, which was operated from January 2003 to May 2007. The HTR-10 operation history provides very important data for the validation of HTGR codes. In this paper, the HTR-10 operation history is simulated with the PANGU code, which has been recently developed for HTGR reactor physics analysis and design. Models and parameters are constructed based on the measured data of the actual conditions. The simulation results agree well with the measurements in all steady-state power periods. The discrepancy of keff is generally below 0.5%, and the discrepancy of coolant outlet temperature is generally below 5°C. It is also figured out that the burnup of graphite impurities has considerable influence on the keff at the end of the operation history, which can cause over 1.5% discrepancy when neglecting the burnup of graphite impurities. By this work, the PANGU code’s applicability in actual HTGR fuel cycle simulations is demonstrated.
The 10 MW High Temperature Gas-cooled Reactor-Test Module (HTR-10)
The HTR-10 operation history provides very valuable data to validate the codes employed in the HTGR analysis and design. Some of these data have been explored as benchmark test cases
The PANGU code
Due to the complexity of the HTR-10 operation history, big efforts have been made to prepare the models and parameters for the simulation work. First, fine time steps are employed in the step-by-step fuel cycle simulation, and the input parameters are processed from the measured data in all detailed power periods. Second, the pebble flow and shuffling model is constructed based on the actual pebble loading and discharging records. Third, the burnup of graphite impurity is considered to overcome the keff discrepancy at the end of operation history. As such, satisfactory simulation results are finally obtained with the PANGU code.
The remainder of this paper is organized as follows.
The scheme of the HTR-10.
Following the FC state, the pebbles in the core started recycling. The graphite pebbles were at first discharged from the fuel discharging tube and the core bottom, which were partially replaced with fresh fuel pebbles and then reloaded into the core. Thus, the ratio of the fuel pebbles in the core was increased gradually. With the progress of pebble recycling, the mixing pebbles also began to be discharged. The first discharged fuel pebble was recorded in April 2005.
The HTR-10 operation history simulated in this work is ranging from August 2002 to May 2007, lasting for about 1,700 days. The corresponding power history is shown in
HTR-10 detailed power history.
Summary of HTR-10 power history (
Year | Power operation time (days) | Integrated power (Mwd) |
---|---|---|
2003 | 106 | 258.9 |
2004 | 168 | 708.5 |
2005 | 149 | 821.4 |
2006 | 97 | 532 |
2007 | 49 | 182.6 |
Total | 569 | 2503.4 |
During the HTR-10 operation history, some important data were measured in details. As for the fuel cycle simulation, the following three categories of measured data need be utilized. The first-category data is the measured control rod position in each power period, as shown in
Control rod position (averaged) during the operation history.
The HTR-10 power history includes a total of 1,020 time periods. A lot of the power periods are quite short, reflecting the transient state of reactor starting up, shutting down, or changing power. Since the PANGU code is mainly used for the steady-state analysis, this work is focused on the simulation results of the steady-state power periods. Nevertheless, in order to conform to the realistic burnup and decay history, all of the detailed 1,020 power periods are explicitly treated by step-by-step fuel cycle simulations with PANGU, without any combination of the short power periods.
Noting that the original thermal-hydraulic data were measured in longer time periods compared with the power data, the thermal-hydraulic input parameters of the fine time steps are calculated by linear interpolation. Because PANGU employs a 2D R-Z model for whole core criticality calculations, it uses an averaged control rod position in each power period, which is calculated from the measured data. Finally, a complete input-parameter table is built for the subsequent simulations, the example data of which is shown in
Input-parameter table used in PANGU simulation (example data).
Step | Time (day) | Time periods (day) | Power (MW) | System pressure (MPa) | Inlet mass flow (kg/s) | Inlet temperature (°C) | Control rod position (cm) |
---|---|---|---|---|---|---|---|
1 | — | — | — | — | — | — | — |
2 | — | — | — | — | — | — | — |
… | — | — | — | — | — | — | — |
1,020 | — | — | — | — | — | — | — |
Pebble flow and shuffling model.
One main challenge in simulating the pebble-bed HTGR operation is related to the treatment of on-line refueling. In the first stages of the HTR-10 operation history, i.e. from the initial core to the full core state, mixing pebbles were loaded and the core height increased along with the reactor operation. In the second stages, the pebbles were recycled through the core and the graphite pebbles were gradually replaced with the fuel pebbles. The two stages are referred as loading stage and recycling stage, respectively.
The pebble flow model in PANGU is improved from the model used in the VSOP code (
As shown in
Pebble-flow model used in PANGU simulation.
Then, a number of shuffling steps are defined in the PANGU fuel cycle simulation. In the beginning of the loading stage, the regions in certain top layers are set as vacuum according to the actual core height of the first time step. With the increase of the loading height, the vacuum regions are filled with mixing pebbles level by level. During the recycling stage, the pebbles flow down along the channel, so that the pebbles of the bottom layer are discharged and the top layer is filled with new loaded pebbles. In each shuffling step, there are 500 mixing pebbles loaded into the core. The mixing ratio of these pebbles are evaluated from the pebble loading data in the operation history, by simply counting the ratio of the loaded fresh fuel, depleted fuel and graphite pebbles in every 500 pebbles.
Mixing ratio of loaded pebbles in the shuffle steps.
Shuffling step | Fresh fuel pebble | Depleted fuel pebble | Graphite pebble |
---|---|---|---|
1 | 0.57 | 0.00 | 0.43 |
2 | 0.57 | 0.00 | 0.43 |
3 | 0.67 | 0.00 | 0.33 |
4 | 0.70 | 0.00 | 0.30 |
5 | 0.70 | 0.00 | 0.30 |
6 | 0.70 | 0.00 | 0.30 |
7 | 0.70 | 0.00 | 0.30 |
8 | 0.70 | 0.00 | 0.30 |
9 | 0.70 | 0.00 | 0.30 |
10 | 0.70 | 0.00 | 0.30 |
11 | 0.71 | 0.02 | 0.27 |
12 | 0.63 | 0.07 | 0.30 |
13 | 0.48 | 0.22 | 0.30 |
14 | 0.50 | 0.24 | 0.26 |
15 | 0.50 | 0.29 | 0.21 |
16 | 0.50 | 0.29 | 0.21 |
17 | 0.27 | 0.41 | 0.33 |
18 | 0.27 | 0.41 | 0.33 |
19 | 0.49 | 0.42 | 0.09 |
20 | 0.35 | 0.59 | 0.07 |
21 | 0.34 | 0.62 | 0.05 |
Besides, based on the recorded number of discharged pebbles and their mixing ratios, the flow speed of each channel can be roughly estimated. For instance, the flow speed of the central channel is estimated according to the time that the first fuel pebble was discharged. In this work, the number of regions in the five channels are 46, 50, 51, 55, and 68 respectively.
In the HTGR graphite, there are dozens of impurity isotopes, such as boron, chlorine, barium, iron, cadmium, and so on, which has considerable influence on keff. In the practical design of HTGR, the graphite impurities are described by the equivalent boron content (EBC)
In our previous study
PANGU adopts a two-step calculation scheme. Burnup and temperature dependent cross-section tables are pre-generated with the lattice code XPZ
The whole fuel cycle simulation contains 1,020 time steps. In each time step, there are iterations between the criticality calculation and the steady-state thermal hydraulics feedback, to obtain the converged keff and temperatures. After that, burnup or decay calculation is performed for the current-step time period. When required, the control rod position and the fuel shuffling is treated at the end of the step. The overall calculation flow is shown in
Calculation flow of the PANGU code simulation.
In order to analyze the influence caused by the burnup of the EBC, comparison simulations are done by changing the burnable ratio to 0 and 100%, respectively. In case of the non-burnable EBC, the graphite impurities (represented by EBC) are not burned during the power operation, which is expected to cause reactivity penalty in the fuel cycle calculation. In contrary, the 100% burnable EBC calculation condition is expected to result in extra reactivity. In
Influence of the EBC burnup
The large uncertainties caused by the nuclear data in HTGR simulations have been reported in some previous work (
The HTR-10 operation history has been simulated with the PANGU code, using delicate models and parameters converted from the measured data. The simulation results are satisfactory, and the PANGU code’s applicability is validated. Future work could be done to investigate the sensitivities and uncertainties caused by the nuclear data and some other input parameters in the simulation. Also, it would be of interest to propose and publish a practical burnup benchmark based on the HTR-10 operation data.
The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.
DS: code development, numerical simulation, and manuscript writing. FC: operation data analysis. BX: operation data preparation, and simulation result evaluation. LS: guidance and consultancy.
This work is supported by the National S&T Major Project (Grant No. ZX06901/ZX06902) of China, and CNNC Youth Research project.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.