Bo Deng

Condition prediction for liquid hydrogen containers is crucial to their safety and efficient utilization. An explainable multi-fidelity machine learning model was established to predict the ventless hold time of 500 L vehicle liquid hydrogen cylinder, and the effects of machine learning algorithms and the proportion of high-fidelity data on model performance were analyzed. The results showed that ensemble learning algorithms exhibited superior accuracy and lower computational overhead compared to the multilayer perceptron. In terms of prediction accuracy, the random forest algorithm achieved the lowest root mean square and mean absolute error on the test set, with values of 4.04 and 2.10. The enhanced low-fidelity data strategy effectively reduced the discrepancies between data from different fidelity levels, while the light gradient boosting machine was the optimal choice to conduct multi-fidelity modeling. The accuracy of the enhanced multi-fidelity model greatly outperformed the multi-fidelity model without enhancement. As the proportion of high-fidelity data in the training set increased from 30% to 35%, the root mean square error of the enhanced multi-fidelity model decreased by 4.7%. In addition, an explainable analysis revealed that pressure increment and heat leakage were dominant promoters affecting ventless hold time. This multi-fidelity modeling study provided a method for predicting ventless hold time that combines accuracy with explainability, and offered insights into the state prediction of liquid hydrogen containers.

As sustainable development becomes increasingly important, technologies for liquid hydrogen (LH₂) storage and transportation are advancing rapidly. Safety concerns regarding LH₂ tanks in fire accidents require further attention. In this study, a one-dimensional thermodynamic model was developed based on layer-by-layer analysis to assess the heat transfer performance of the insulation structure in LH₂ tanks under fire conditions. Through the transformation of the solving target and iteration rules, a novel and efficient solution method was proposed for such thermodynamic problems. The thermodynamic performance of the insulation structure coupled with spray-on foam and variable-density multilayer under normal temperature (NT) and standardized fire conditions (863.15 K) was analyzed, and the effects of insulation structure parameters and environmental factors were evaluated. A case study of a 500 L vehicle LH₂ tank was conducted using the software package BoilFAST, with the total heat leakage as the key input, to analyze the evolution of internal pressure and temperature. Results show that within the insulation structure, temperature decreases rapidly by 80.35% and 89.55% under fire and NT conditions, respectively. Spray-on foam insulation thickness, layer density, residual gas pressure, and hydrogen temperature exert minor effects, while the temperature of the external environment and the number of layers significantly affect the heat flux under the fire condition. Under the NT condition, heat leakage is primarily attributed to support structures and accessory pipelines, whereas under the fire condition, heat leakage from the insulation structure becomes the main source, accounting for 63%. This study provides a reference for heat transfer assessment of LH₂ tanks in fire scenarios.

The full multilayer high-pressure hydrogen storage vessel plays an important role in hydrogen refueling stations. However, these vessels may fail after a certain period due to crack formation, necessitating periodic inspections. Among the various parts, the butt joints connecting the thick-walled nozzles and hemispherical heads represent critical and challenging areas for inspection. In this study, a one-shot multi-receiver defect detection and localization method is developed based on the ultrasonic phased array method. In order to verify the feasibility of the method, the interaction between the ultrasonic wave and the crack defects at the key position of the butt joint is analyzed based on finite element, enabling the accurate localization of crack tips; an experimental specimen was designed and fabricated, and a corresponding phased array detection test was conducted to validate the method.

Metal oxide semiconductor (MOS) hydrogen sensors offer advantages, such as high sensitivity and fast response, but their challenges remain in achieving low-cost fabrication and stable operation at room temperature. This study investigates Nb-doped TiO₂ (NTO) thin films prepared via a one-step micro-arc oxidation (MAO) with the addition of Nb₂O₅ nanoparticles into the electrolyte for room-temperature hydrogen sensing. The characterization results revealed that the incorporation of Nb₂O₅ altered the film’s morphology and phase composition, increasing the Nb content and forming a homogeneous composite thin film. Hydrogen sensing tests demonstrated that the NTO samples exhibited significantly improved sensitivity, selectivity, and stability compared to undoped TiO₂. Among the fabricated samples, NTO thin film prepared at Nb₂O₅ concentration of 6 g/L (NTO-6) showed the best performance, with a broad detection range, excellent sensitivity, rapid response, and good specificity to hydrogen. A strong linear relationship between response values and hydrogen concentration (10–1000 ppm) highlights its potential for precise hydrogen detection. The enhanced hydrogen sensing mechanism of NTO thin films primarily stems from the influence of Nb₂O₅; nanoparticles doping in the anatase-phase TiO₂ structure on the semiconductor surface depletion layer, as well as the improved charge transfer and additional adsorption sites provided by the Nb/Ti composite metal oxides, such as TiNb₂O₇ and Ti_0.95Nb_0.95O₄. This study demonstrates the potential of MAO-fabricated Nb-doped TiO₂ thin films as efficient and reliable hydrogen sensors operating at room temperature, offering a pathway for novel gas-sensing technologies to support clean energy applications.