Accurate prediction of transient three-dimensional temperature fields during organ cooling and rewarming is essential for maintaining viability in cryopreservation. However, this task is fundamentally constrained by the scarcity of experimentally measured data, the distribution discrepancy between simulated and real experimental data, and the strong nonlinearities associated with phase-transition processes. This paper proposes a Transient Temperature Query Network (TTQN) that integrates simulated and real experimental data by reformulating temperature-field reconstruction as a query-driven conditional regression task, and establishes a large-scale mixed dataset covering diverse animal tissues and organs. By incorporating multi-channel thermophysical parameters, multi-scale feature-extraction modules, and physics-guided loss functions, the proposed model enables precise temperature prediction at arbitrary internal positions of organs under sparse observation conditions. The proposed TTQN demonstrates superior prediction performance compared with several benchmark models, achieving stable reconstruction accuracy under limited observation conditions and significantly reducing simulation-to-experiment discrepancies. This study establishes a scalable and generalizable paradigm for three-dimensional temperature-field prediction in biomaterials, providing core technical support for precise thermal-regulation strategies during organ preservation.
Variable-orientation thermal management is increasingly required in embodied-intelligence robots and aerospace electronics, where changes in gravity orientation can alter vapor–liquid distribution and compromise two-phase cooling stability. Although the distributed jet array shortens vapor-discharge paths and enhances high-heat-flux cooling, the effects of gravity orientation on boiling hysteresis, vapor blockage, and critical heat-transfer performance remain unclear. This study constructed a rotatable R245fa mechanically pumped two-phase loop integrating an embedded distributed jet array boiling module and a miniature test chip with multipoint in situ thermometry. Complete boiling curves were obtained for both the increasing and decreasing heat flux branches at different gravity inclination angles (0.2°–180.0°) and jet Reynolds numbers (4728–7880), using synchronized thermal–hydraulic measurements and high-speed visualization. Boiling hysteresis was governed mainly by the orientation-sensitive onset of nucleate boiling, whereas the boiling extinction point was comparatively insensitive to orientation, producing a non-monotonic response. By contrast, critical heat flux decreased monotonically with increasing inclination angle. As the angle increased from 0.2° to 180.0°, critical heat flux decreased by 48.0%, 39.1%, and 25.5% at Reynolds numbers of 4728, 6304, and 7880, respectively, suggesting that higher jet inertia may reduce orientation sensitivity. Orientations unfavorable for vapor removal also narrowed the stable nucleate boiling window and promoted gravity-biased vapor accumulation, intermittent vapor blockage, and pressure fluctuations. Correlations were developed for the two-phase heat transfer coefficient and critical heat flux, with a gravity-orientation factor incorporated to account for orientation-dependent vapor removal and liquid rewetting. The corresponding mean absolute percentage errors were 2.89% and 3.76%, respectively. These results reveal stage-dependent orientation effects and provide experimental data and empirical tools for evaluating the operating limits and orientation tolerance of R245fa mechanically pumped two-phase cooling devices.
Oxygen-enriched combustion along with fast-responded intelligent control has a key role to play in clean operation of modern industrial furnaces. However, there is lack of a model for estimating flue gas radiation emissivity, which give a right balance between computational efficiency and prediction accuracy. Present radiation models such as line-by-line (LBL) model have a high accuracy but with an unacceptable computation cost, while weighted-sum-of-gray-gases (WSGG) model is the reverse. Herein, a computationally efficient Layer-over-layer model (LOL) without sacrificing accuracy is developed for radiative energy calculation based on "spatial accumulation after layered spectral modeling". Through modeling at temperatures of 600–2000 K, pressures of 0.1–65.0 bar, path-lengths of 0.1–20.0 m, and H2O(g)/CO2 ratio of 2:1, the LOL model was found to be as accurate as the LBL model with a maximum error below 2.0%, and computationally efficient as the WSGG model with some four orders of magnitude quicker than the LBL. The flue gas emissivity was found to be logarithmically dependent on radiative molecule number, total pressure (P) and the H2O(g)/(H2O(g) + CO2) ratio, while following a Gompertz-type dependence on temperature (T). An empirical formula for emissivity calculation was derived using nonlinear least squares regression, linking with the radiative molecule number, pressure, composition, and temperature. The empirical formula is capable of estimating chamber height for typical re-heating furnaces with oxy-enriched combustion. With 100% oxygen concentration at P = 1 bar and T = 1600 K, the H2O/CO2 ratio in flue gas is 2:1, and the flue gas layer thickness (a key parameter related to furnace height) can be reduced by 76.4% compared to that with conventional air-fueled combustion.
The increasing package-level integrations of high-power-density electronics, combined with system-level mounting constraints in practical hardware platforms, are driving the need for large-footprint heat sinks, which can maintain high cooling performance in both horizontal- and vertical-mounting configuration. However, the flow boiling behavior under vertical-mounting configuration, with a vertically oriented heated surface and horizontally aligned flow channels, is still unclear. In this study, flow boiling of R134a was experimentally investigated in a large-footprint (68 mm × 80 mm) heat sink under vertical-mounting conditions at mass fluxes of 103 and 180 kg/m2·s, heat fluxes of 17–207 kW/m2, and an inlet subcooling of 10 °C. Temperature measurements and high-speed flow visualization were employed to examine the effects of buoyancy, outlet-port arrangement, and parallel channel instability (PCI) on vapor transport and rewetting dynamics. The results reveal that buoyancy-induced vapor accumulation is the primary cause of local dryout and thermal nonuniformity, producing temperature differences of up to 11.3 °C across the heated surface. Relocating the outlet port toward the upper region of the heat sink facilitates vapor discharge along the buoyancy direction and partially alleviates vapor accumulation. More importantly, PCI was found to induce periodic flow redistribution and accelerated liquid-wave propagation, resulting in intermittent rewetting of dryout regions and suppression of vapor blanketing. High-speed visualization directly captured the PCI-assisted rewetting process and showed that PCI could self-regulate to variations in heat flux, plenum size, mass flux and outlet port configuration, thereby maintaining sufficient local mixture velocity for vapor removal and reducing temperature nonuniformity by 7 °C. The results provide new insights into the interplay between flow instability and rewetting, while establishing that PCI, traditionally regarded as detrimental, can play a beneficial role in large-footprint two-phase cooling systems under vertical-mounting configurations.
Temperature and flow rate are critical operating parameters that affect the performance and bubble behavior of the anion exchange membrane water electrolyzer (AEMWE), directly influencing mass transfer efficiency and operational stability. To elucidate their underlying mechanisms, an AEMWE with transparent anode end plates was designed to enable in situ observation of bubble dynamics within the anode flow field. By combining electrochemical impedance spectroscopy (EIS) measurement and distribution of relaxation times (DRT) analysis, the performance and bubble characteristics under different temperature–flow rate combinations were systematically investigated. The results indicated that elevating the temperature improved electrolytic performance with minimal impact on bubble coverage. While increasing the flow rate significantly reduced bubble coverage, it concurrently lowered the cell temperature, negatively affecting electrolytic performance. Based on experimental findings, this study reveals the synergistic relationship between temperature and flow rate, offering a physically plausible interpretation that the convective cooling effect at higher flow rates may dominate over the benefits of macroscale bubble removal. It proposes an optimized operational strategy: maintaining the system within an efficient temperature range through coordinated temperature control while keeping flow rate within an appropriate range. This research provides insights for optimizing operating conditions and ensuring efficient, stable operation of AEMWE systems.
To deeply explore the energy-saving potential of a proton exchange membrane fuel cell (PEMFC) system, this study established thermodynamic models of system components based on actual system operational data. Building upon conventional exergy analysis to clarify exergy destruction, this employs advanced exergy analysis methods to deconstruct total exergy destruction into avoidable/unavoidable and endogenous/exogenous segments. This approach not only characterized the location and magnitude of the destruction, but also achieved a profound diagnosis of the roots of irreversibility and the potential for optimization. Research findings indicate that 66.55% of the total exergy destruction in the system constitutes avoidable destruction, highlighting significant energy-saving opportunities. The loss mechanisms of key components are distinct: the stack destruction is composed of both unavoidable fundamental irreversibility and avoidable polarization destruction. In contrast, air supply system exergy destruction predominantly stems from the inefficiency of components (endogenous), while the destruction in the cooling system is mainly caused by system-level integration constraints (exogenous). Consequently, this study delineates differentiated optimization pathways: for the stack, the focus should be on mitigating its avoidable polarization destruction through material and design innovations; for the air supply system, component technological upgrades are imperative to reduce its avoidable-endogenous destruction; and for the cooling system, effort must be directed toward system-level coordinated control optimization to overcome its avoidable-exogenous destruction. This research provides a theoretical foundation and decision-making support for enhancing the performance of automotive fuel cell systems.
The macropore-matrix system plays a critical role in governing preferential flow under field conditions, and understanding solute transport within such a system forms the basis for investigating the complex fate of contaminants in heterogeneous porous media. Moreover, single macropores in the field are often partially or completely filled with sediments, yet the process of reactive transport in such filled macropore–matrix systems has received limited attention. A mobile–immobile (MIM) model is developed for sediment-filled macropore–matrix systems, representing solute transport as a three-domain process with advection, radial dispersion, reactions, sorption, and rate-limited mass exchange. The governing equations are solved using Laplace-domain analysis with numerical inversion and validated against numerical simulations. In addition, Markov Chain Monte Carlo (MCMC) is applied to estimate the MIM model parameters using the experimental data. Results from the semi-analytical solution show that pore filling strongly modifies solute transport and breakthrough behavior. A reduction in the porosity ratio between the mobile and immobile regions lowers the peak of the breakthrough curves (BTCs) and amplifies tailing. Likewise, a smaller mass-transfer coefficient produces higher pore concentrations and modifies BTC tailing by delaying mass exchange between the mobile and immobile regions. Analysis of diffusion fluxes further reveals that back-diffusion is an important mechanism responsible for the observed tailing behavior. Parameter inversion using the MCMC method based on the proposed model shows a marked improvement over existing approaches, reducing root mean square error (RMSE) by up to 55% in the column experiment. Overall, the proposed model effectively captures the solute transport characteristics in macropore–matrix systems and provides valuable theoretical and practical insights for groundwater pollution studies in heterogeneous aquifers.
Q235B low-carbon steel is widely used in engineering fields due to its excellent comprehensive properties and cost advantages. However, its high-temperature oxide scale significantly compromises the quality of subsequent welding and coating processes. This study proposes a continuous‑pulsed hybrid laser method for oxide scale removal, with a focus on investigating the influence of the time interval Δt between the continuous‑wave and pulsed lasers on removal quality and the underlying mechanisms. Combining numerical simulation and experimental validation, the study systematically analyzes the oxide removal mechanisms, surface morphology evolution, microstructure, and nanohardness changes under different Δt. The results suggest a threshold effect: when Δt ≤ 10 ms, plasma shielding induced by the continuous laser weakens removal efficiency; when Δt ≥ 15 ms, the shielding is substantially reduced and the continuous laser's preheating effect enhances the material's absorptivity to pulsed laser energy, promoting spattering and increasing removal amount. The optimal Δt = 15 ms achieves the lowest surface oxygen content and highest removal amount. Under this condition, surface grains are significantly refined with a 58.8% reduction in average grain size and increased GND density, leading to a 51.5% enhancement in nanohardness from 3.30 GPa to 5.00 GPa. This work elucidates the critical role of the time interval in hybrid laser removal, providing theoretical and experimental foundations for efficient, high-quality oxide scale removal and surface property enhancement on low‑carbon steel.
Synthetic jet (SJ) impingement cooling is a promising technique for enhancing convective heat transfer in compact thermal management systems. While the behaviour of single jets is well understood, practical applications require multiple jets operating in close proximity. In such configurations, strong nonlinear interactions between neighbouring jets significantly alter the flow structure and cooling performance, making accurate prediction challenging. This study develops an integrated computational, experimental, and machine-learning framework to predict the thermal behaviour of multi-SJ arrays. Transient CFD simulations, validated using hot-wire anemometry, are employed to characterise jet dynamics and generate a dataset spanning key non-dimensional parameters, including Reynolds number (Re), the jet-to-surface spacing (H/D), lateral spacing (S/D), stroke length ratio (L/D), and actuation frequency. Conventional data-driven models trained on single-jet data fail to generalise to multi-jet configurations, while models trained on combined datasets remain restricted to fixed geometries. To overcome this limitation, a Modular Jet-wise Reconstruction framework is proposed, in which local single-jet predictions are superposed and augmented with an interaction-aware correction model to capture nonlinear jet-jet effects. The proposed framework enables accurate prediction across the investigated range of varying inline jet configurations without retraining, while achieving a computational speed-up of approximately 107 compared to transient CFD. This provides a computationally efficient framework for multi-jet cooling within the investigated configuration and parameter ranges and enables rapid design exploration.
