یافتن بهترین مجموعه های سوخت برای زدن طرح ADS جهت تغییر شکل MA با ​​استفاده از برنامه ریزی پویا

Dynamic Programming is applied to find the best shuffling scheme of ADS, which consists of 54 fuel assemblies and a spallation target of Pb–Bi, and the best scheme is successfully found under different constraints of power peaking factor. In the present problem the average transmutation ratio of MA is maximized under the constraints for effective neutron multiplication factor, discharged fuel burn-up, and power peaking factor. The obtained results show the best fuel shuffling scheme consists of initial unique shuffling pattern and succeeding repetitive patterns for all cases. The repetitive pattern is unique for each constraint of power peaking factor, the best shuffling scheme changes sensitively for the change of constraints. For most best shuffling schemes, fresh fuel is loaded into the 2nd inner row of fuel region, and finally discharged from the inner or outer region because of the constraint of power peaking factor.

Several countries recently promote research and development of accelerator-driven system (ADS) to reduce the burden for conditioning and disposal of the high-level radioactive waste (HLW) by transmuting minor actinide (MA) [1]. The ADS is a hybrid system that consists of a high intensity accelerator, a spallation target and a subcritical reactor. In usual ADS designs [2] and [3], spallation neutrons are supplied at the center of its core. In most designs lead–bismuth eutectic (LBE) is used as a spallation target, which is also used as a coolant. This kind of ADS generally has a higher power peak than critical reactors because of its strong neutron source at the center. The power peak may cause the violation of temperature limit of the fuel cladding tubes. Therefore, it is important to flatten the power density distribution. The ADS is a subcritical reactor system and considered safe for a reactivity-initiated accident by an enough reactivity margin. On the other hand, it is necessary to adjust a power level of the accelerator for supplementing the subcriticality of ADS. The larger k-eff is better from economical viewpoint since the larger power can be obtained even for smaller accelerator power, though its safety margin becomes smaller. In the previous research for ADS by Japan Atomic Energy Agency (JAEA) [3], the subcritical core is composed of several regions and each region has different fuel composition in order to make the power density distribution flat. In this case, it is necessary to fabricate many kinds of fuel pellets. The fabrication of the ADS fuel is difficult because main component of this fuel is MA and Pu. The multiregion design using various fuel compositions is a burden for ADS operation. In the present study the multiregion design using fuel shuffling is adopted to solve this problem. It is expected that the fuel shuffling can make the power distribution flat and transmute MA effectively by using only one kind of fuel composition. However, the investigation of the ADS fuel shuffling has been hardly done. Therefore in this study we try to find the best fuel shuffling scheme for hexagonal small ADS and investigate regularity of this scheme. There are a lot of patterns of how to shuffle fuels, so the total number of calculation cases becomes huge for finding the best scheme. In this study the Dynamic Programming [4] which can reduce the amount of calculation is introduced.

Dynamic Programming has been successfully applied to find the best shuffling scheme of a 470 MWth ADS, which consists of fuel of (MA + Pu)N, inert matrix of ZrN, and spallation target and coolant of Pb–Bi. Average transmutation ratio of MA is used as the performance index, and the maximum discharged fuel burn-up, range of k-eff and power peaking factor are chosen as the constraints. The obtained results show the best fuel shuffling scheme consists of initial unique shuffling pattern and repetitive pattern for all cases. It is found that the shuffling scheme changes sensitively for different constraint value of power peaking factor. However transmutation rate of MA is about the same if the number of loaded fresh fuels is same, therefore the number of fresh fuel is important for increase of transmutation rate of MA. For most best shuffling schemes, the fresh fuel is loaded into the 2nd row of fuel region, and finally discharged from the inner and outer region.