Deep neural operator-driven real-time inference to enable digital twin solutions for nuclear energy systems.

This paper focuses on the feasibility of deep neural operator network (DeepONet) as a robust surrogate modeling method within the context of digital twin (DT) enabling technology for nuclear energy systems. Machine learning (ML)-based prediction algorithms that need extensive retraining for new reactor operational conditions may prohibit real-time inference for DT across varying scenarios. In this study, DeepONet is trained with possible operational conditions and that relaxes the requirement of continuous retraining - making it suitable for online and real-time prediction components for DT. Through benchmarking and evaluation, DeepONet exhibits remarkable prediction accuracy and speed, outperforming traditional ML methods, making it a suitable algorithm for real-time DT inference in solving a challenging particle transport problem. DeepONet also exhibits generalizability and computational efficiency as an efficient surrogate tool for DT component. However, the application of DeepONet reveals challenges related to optimal sensor placement and model evaluation, critical aspects of real-world DT implementation. Addressing these challenges will further enhance the method's practicality and reliability. Overall, this study marks an important step towards harnessing the power of DeepONet surrogate modeling for real-time inference capability within the context of DT enabling technology for nuclear systems.

Neutronic investigation of alternative & composite burnable poisons for the soluble-boron-free and long life civil marine small modular reactor cores.

Concerns about the effects of global warming provide a strong case to consider how best nuclear power could be applied to marine propulsion. Currently, there are persistent efforts worldwide to combat global warming, and that also includes the commercial freight shipping sector. In an effort to decarbonize the marine sector, there are growing interests in replacing the contemporary, traditional propulsion systems with nuclear propulsion systems. The latter system allows freight ships to have longer intervals before refueling; subsequently, lower fuel costs, and minimal carbon emissions. Nonetheless, nuclear propulsion systems have remained largely confined to military vessels. It is highly desirable that a civil marine core not use soluble boron for reactivity control, but it is then a challenge to achieve an adequate shutdown margin throughout the core life while maintaining reactivity control and acceptable power distributions in the core. High-thickness ZrB2 150 µm Integral Fuel Burnable Absorber (IFBA) is an excellent burnable poison (BP) candidate for long life soluble-boron-free core. However, in this study, we want to minimize the use of 150 µm IFBA since B-10 undergoes an (n, α) capture reaction, and the resulting helium raises the pressure within the plenum and in the cladding. Therefore, we have considered several alternative and novel burnable BP design strategies to minimize the use of IFBA for reactivity control in this study: (Case 1) a composite BP: gadolinia (Gd2O3) or erbia (Er2O3) with 150 µm thickness ZrB2 IFBA; (Case 2) Pu-240 or Am-241 mixed homogeneously with the fuel; and (Case 3) another composite BP: Pu-240 or Am-241 with 150 µm thickness ZrB2 IFBA. The results are compared against those for a high-thickness 150 µm 25 IFBA pins design from a previous study. The high-thickness 150 µm 25 IFBA pins design is termed the "IFBA-only" BP design throughout this study. We arrive at a design using 15% U-235 fuel loaded into 13 × 13 assemblies with Case 3 BPs (IFBA+Pu-240 or IFBA+Am-241) for reactivity control while reducing 20% IFBA use. This design exhibits lower assembly reactivity swing and minimal burnup penalty due to the self-shielding effect. Case 3 provides ~10% more initial (beginning-of-life) reactivity suppression with ~70% less reactivity swing compared to the IFBA-only design for UO2 fuel while achieving almost the same core lifetime. Finally, optimized Case 3 assemblies were loaded in 3D nodal diffusion and reactor model code. The results obtained from the 3D reactor model confirmed that the designed core with the proposed Case 3 BPs can achieve the target lifetime of 15 years while contributing to ~10% higher BOL reactivity suppression, ~70% lower reactivity swings, ~30% lower radial form factor and ~28% lower total peaking factor compared to the IFBA-only core.