Mingjun Wang

Inertial particle transport and deposition in helically forced wall-bounded turbulence are controlled by turbulent diffusion, near-wall shear, secondary motion, and finite particle inertia. The wire-wrapped rod bundle is treated as a representative configuration in which spacer wires generate cross-flow, local acceleration, and narrow-gap transport pathways. A discrete-phase model with two-way momentum coupling was established to simulate Fe3O4 particle migration and deposition in a 19-pin wire-wrapped rod-bundle flow with lead-bismuth eutectic as the carrier fluid. The continuous phase was resolved using the shear stress transport k–ω model, and a particle–wall collision and deposition criterion was used to distinguish rebound from deposition. Particles of 1–100 μm were tracked to analyze axial migration, transverse redistribution, near-wall enrichment, and wall deposition. Results show that inlet velocity gradients and wire-induced cross-flow drive particles away from streamlines and toward near-wall low-speed regions. For 1–10 μm particles, 61.8% of the area-integrated wall deposition occurred within the first 0.5 wire pitch. Deposition hotspots occur mainly on upstream-facing wire surfaces, box-wall edges, and narrow gaps among the wire, cladding, and box wall. Increasing particle diameter raises relaxation time and Stokes number, enhancing trajectory deviation and collision probability while shifting concentration and deposition-flux peaks downstream. As the particle size increases, the dimensionless deposition rate increased and showed a sublinear dependence on the Stokes number, while the deposition rate remained far below unity, indicating that deposition is governed by coupled turbulent diffusion, near-wall shear, wire-induced transverse migration and finite inertia rather than inertia alone.

Small-scale closed Brayton-cycle systems require compact air-cooled heat exchangers that provide high heat-rejection capacity within strict mass and volume constraints while limiting auxiliary air-moving power. Additive manufacturing expands the topology space beyond conventional manufacturing, but internal overhangs remain constrained by process-specific support requirements. A tree-like self-supporting fin (TSF) is proposed by embedding a nominal laser powder-bed-fusion 45° branch-axis overhang rule into topology generation. The two-level branched fin is screened against this geometric rule, a literature-informed lower feature-size bound and open-flow-space requirements, establishing a self-support-oriented configuration for CFD assessment. Its thermal-hydraulic performance is compared with that of a conventional annular fin using a local-unit CFD model subjected to spatial mesh-sensitivity assessment and canonical correlation comparisons. Within the same envelope, the modeled TSF enlarges the heated area by about 1.8-fold and raises the LMTD-based heat-transfer coefficient by about 3.1-fold at a 2.6-fold static-pressure-drop penalty, giving a PEC-type index of about 2.28 at matched inlet velocity. At matched Pₛ = ΔpₛV̇, used as a local-unit static-pressure flow-power proxy, the interpolated heat-transfer rate is about 3.9 times that of the annular baseline at the reference operating point, and the modeled advantage persists over 2–12 m·s⁻¹. These results establish the air-side convective potential of the TSF under controlled constant-property, isothermal-wall conditions and provide a basis for system-coupled and experimental development of compact Brayton-cycle heat rejection.

Critical heat flux (CHF) is a key thermal limit for boiling heat transfer and reactor safety, but data-driven CHF models often suffer from non-physical extrapolation near boiling crisis boundaries. In this study, a unit-aware physical embedding Transformer is proposed for CHF prediction. The model integrates learnable unit embeddings, unit-interaction features, multi-scale extraction and physics-informed regularization to preserve the physical semantics and scale consistency of thermal–hydraulic variables. The model was evaluated on 24,579 samples from the public CHF tube database under full-range testing and two extrapolation scenarios: high-pressure holdout and high-quality holdout. In the full-range test, the proposed model achieved an RMSE of 213.91 kW·m⁻² and an R² of 0.9818. Compared with dimensionless inputs, the original physical variables provided a more informative representation of pressure, flow, geometry and phase-evolution effects. Under extrapolation conditions, the proposed framework maintained stable performance in high-pressure regimes and achieved the best overall accuracy in the high-quality dryout-dominated regime. Unit-perturbation tests further showed that the model reduced systematic prediction drift by more than 60%. Interpretability analyses indicated physically consistent responses associated with mass flux, pressure and outlet quality. These results demonstrate that embedding unit semantics and physical constraints into Transformer-based CHF modeling can improve the robustness and credibility of thermal-limit prediction under extrapolative boiling crisis conditions.

Helium-Xenon (He-Xe) gas mixtures have emerged as superior working fluids for advanced energy conversion systems. However, the flow and heat transfer of He-Xe mixtures in complex dense heated rod bundles are not well characterized, particularly regarding gravity-driven secondary flows and temperature redistribution effects. In this study, He-Xe physical models, including thermophysical properties, turbulence models, wall functions, and turbulent Prandtl number models are first validated using experimental data to develop an optimized flow and heat transfer analysis approach. Subsequently, building upon the validated models, the thermal-hydraulic parameters in a 7-rod bundle are evaluated for both detailed and simplified wire-wrap configurations. Subsequently, the simplified model is applied to study the flow behavior and temperature redistribution in a larger 61-rod bundle. The results indicate that the Stratification number (STR) within the rod bundle channel correlates negatively with inlet velocity and positively with assembly power. At high flow rates or low heat fluxes, STR exhibits distinct axial periodic oscillations, peaking at 0.55 m. A sharp dip in the Richardson number (Ri) at the 0.55 m axial location is driven by localized heat transfer enhancement. Under pronounced temperature redistribution, intense secondary flows elevate the global Nusselt number (Nu), yet local heat transfer in the upper region of the channel deteriorates; specifically, at an inlet velocity of 0.25 m·s-1, the maximum temperature within the channel increases by 67.3 K. This study can provide reference value for thermal safety analysis of He-Xe cooled energy conversion system.

According to the characteristics of multi-variable control object in small modular reactors control system, a design method of control system is presented in this paper. Firstly, the nonlinear differential equations of the system are constructed according to the principle of mechanism modeling, and then the transfer function of the system is obtained by linearization. Then, the coupling characteristics of the system are analyzed by using Gershgorin Circle Theorem in modern frequency domain analysis method. On this basis, the inverse Nyquist array method is used to decouple the two key channels of reaction-exit temperature channel and feed-water valve-steam pressure in small modular reactors. Then the PI correction network is designed by frequency response method. In order to fully verify the performance of the decoupling controller, the system is verified based on linear model and nonlinear model respectively, and the pre-compensation matrix processing strategy under variable load conditions is proposed. Simulation results show that this method can significantly reduce the coupling degree between the two channels, and the performance of the control system after decoupling is significantly improved compared with that before decoupling.

The forced convective heat transfer capacity of fluid-solid conjugate heating surfaces is closely related to the evolution behavior of the fluid flow field. The strong shear flow induced by complex non-canonical structures can promote the generation of vortices, thereby enhancing the heat transfer capacity and improving the safety margin of the heat exchanger. In recent years, helical cruciform fuel (HCF) has received extensive attention in the design of small modular reactors due to its high-power density, compact arrangement, and complex non-canonical structure. To investigate the refined swirling flow field and mixing behavior under complex twisted structure, based on the validation results of the mixing experiment, flow resistance experiment, and applicability analysis of turbulence models, the computational fluid dynamics (CFD) simulation on a 19-rod hexagonal-arranged dense-lattice HCF assembly was conducted. The causes of strong shear flows and swirling vortex distributions in the rod bundle flow field were analyzed. Meanwhile, quantitative analyses of transverse mixing and flow resistance characteristics of the HCF assembly with different flow rates were carried out. Different from cylindrical rods, the mixing coefficient of the helical fuel doesn't exhibit a monotonic variation trend with Reynolds number; instead, it mainly manifests as forced flow sweeping along complex twisted structures. The average mixing parameters of the central, side, and corner channels are 0.028, 0.027, and 0.04, respectively. Furthermore, the flow resistance model of HCF assembly based on Rehme's theory and a refined mixing model is developed, with an average prediction accuracy improved by 44.8%.