The increasing penetration of renewable energy necessitates efficient and site-independent energy storage solutions to ensure grid stability. Liquid air energy storage (LAES) has emerged as a promising candidate owing to its high energy density and absence of geographical constraints, while its round-trip efficiency remains limited by the energy-intensive air liquefaction process. This study proposes novel air liquefaction cycles integrated with vapor-injection refrigeration (VIRS) as an inter-stage precooling strategy to enhance the efficiency and compactness of microgrid-scale LAES systems. A comprehensive parametric analysis using Aspen HYSYS evaluates the integration of VIRS with Linde-Hampson, Claude, and Kapitza cycles, examining key variables including air compression pressure (pcomp,air), VIRS evaporation temperature (Tevap) and expander inlet temperature (Texp,in). Results demonstrate that VIRS precooling substantially enhances liquefaction performance across all cycles. The integration of VIRS precooling with Linde-Hampson, Claude and Kapitza cycles achieves the optimal SEC of 1.19 kWh/kg, 0.493 kWh/kg and 0.461 kWh/kg with the reduction ratios of 54.62%, 16.19% and 15.03% compared with the conventional air liquefaction cycles without precooling, while the counteractions among the parameters of VIRS and air process are comprehensively evaluated. Advanced exergy analysis quantifies the thermodynamic irreversibility and confirms the crucialness of air expander for mitigating exergy waste, while the precooling energy input contributes to substantial exergy savings within the air liquefaction process. These findings establish a foundation for developing distributed, miniaturized LAES systems with enhanced thermodynamic and economic viability for microgrid applications.

Measurement of the gas flow is required during the transportation process of hydrogen-blended natural gas. Ultrasonic flow meters are widely used in oil and gas pipeline transportation metering. The presence of hydrogen-blended natural gas components, particularly multiphase flow components, can affect the accuracy of in-service ultrasonic metering systems. The study uses the simulation software COMSOL Multiphysics (R) to investigate the propagation characteristics of simulated ultrasonic signals in water-containing hydrogen-blended natural gas pipelines. This aim of this study is to explore the influences of factors such as water content, transducer installation angle, and transmitted signal frequency on ultrasonic propagation characteristics in water-containing hydrogen-blended natural gas. The results show that the water content in the pipelines affects the propagation speed of ultrasonic waves. The transducer was installed at the most suitable angle, 45 degrees. A higher transmission frequency increases the instantaneous intensity of ultrasonic waves in pipelines, making the signal easier to detect. However, it will cause energy dispersion, reducing measurement accuracy. The study's result can provide a theoretical basis for measuring the flow of water-containing hydrogen-blended natural gas using ultrasonic flowmeters.

Efficient mixing of hydrogen-blended natural gas is critical for safe and large-scale hydrogen transportation. This study proposes a vortex generator-based static mixer to enhance hydrogen dispersion by strengthening the hydrogen jet and inducing counter-rotating vortices. Five structural parameters, including angle of attack, height, trailing-edge chord length, number of vortex generators, and hydrogen inlet diameter, were optimized using orthogonal experimental design, entropy weight method, and Technique for Order Preference by Similarity to Ideal Solution. The optimal configuration was obtained at an angle of attack of 90°, height of 300 mm, trailing-edge chord length of 200 mm, three vortex generators, and hydrogen inlet diameter of 120 mm. Compared with a conventional Kenics static mixer, the proposed mixer reduced the mixing lengths required for 90% and 95% hydrogen concentration uniformity by 3.5 m and 4.3 m, respectively, while decreasing pressure drop by 3556 Pa. These results demonstrate its superior mixing efficiency and lower flow resistance.

Sound source localization using deep learning presents great potential, but its widespread application is often hindered by the limited availability of real-world labeled data needed to train high-capacity models. This work introduces a deep learning-based framework designed to address this data scarcity challenge in indoor direction-of-arrival (DoA) estimation tasks. Specifically, complex Morlet wavelet transforms are used to generate high-resolution time-frequency representations from multichannel microphone array signals, capturing both temporal and spectral information, including crucial phase cues. These representations are then fed into a hybrid CoAtNet architecture that combines convolutional layers with self-attention mechanisms to enable effective local feature extraction and global spatial context modeling. To mitigate the dependence on extensive real-world datasets, a two-stage training strategy is adopted: large-scale synthetic data generated via Pyroomacoustics is used for pretraining, followed by fine-tuning on a small set of real-world samples for domain adaptation. Experimental results demonstrate that the proposed system achieves 98.22 % accuracy on real recordings and 95.21 % on the SLoClas benchmark dataset, outperforming baseline deep learning models. The proposed framework offers a practical and efficient solution for sound source localization in real-world applications where labeled data is limited.

Leveraging existing natural gas pipelines to transport hydrogen-blended natural gas is a practical route for largescale hydrogen delivery. Ultrasonic flowmeters are essential for pipeline metering, but residual water vapor may condense into droplets on the transmitting surface of the transducer, disturbing ultrasonic propagation and energy distribution. In this study, a multiphysics model was developed in COMSOL Multiphysics to investigate the effects of attached droplets on the acoustic and flow fields in hydrogen-blended natural gas pipelines. The influences of droplet diameter, emission frequency, hydrogen blending ratio, and droplet position were systematically analysed. The results show that droplet diameter has a significant attenuation effect on the received signal. As the droplet diameter increased from 1 to 4 mm, the peak acoustic intensity at the receiving end decreased from 2.69 & times; 105 to 2.59 & times; 102 W/m2. Increasing the emission frequency from 15 to 60 kHz shortened the acoustic-wave arrival time from 550 to 380 mu s, but intensified wave interference. When the hydrogen blending ratio increased from 5 % to 20 %, the average acoustic energy density decreased from 8.42 to 6.88 J/ m3. Moreover, off-centre droplets weakened the influence on arrival time but still affected local instantaneous acoustic intensity. These findings guide ultrasonic flowmeter deployment and signal correction in hydrogenblended natural gas pipelines.

In this paper, the adsorption properties of NO2 on W-modification ZnO, based on the synergistic effect of Wolframium (W) and Oxygen vacancy defects, has been investigated using density functional theory (DFT) calculations. The modulation of hypervalent transition metal W greatly changed the electronic structure of the crystal plane, which could promote the generation of oxygen vacancy defects on the crystal plane. Under the synergistic effect of W-doped and oxygen vacancies, the conductivity of the ZnO (002) could be effectively enhanced, that would mean the improvement of NO2 adsorption. Energy analysis shows that the adsorption energy is improved from the original -0.752 eV to -7.506 eV, a 9.98-fold enhancement. Mulliken charge population analysis shows that the charge transfer amount increased from -0.316 e to -0.941 e, which is 2.97 times higher. Theoretical calculations show that the sensitivity in adsorption of NO2 can reach 18.626, which is about 24.4 times that of the intrinsic one. Moreover, the W-doped ZnO material exhibits superior adsorption selectivity for NO2 compared to other gases such as CO, CO2, H2, and NO, which can provide new ideas for the design of NO2 gas sensors with ultra-low concentrations.

To enhance the compression efficiency and refrigerant flow capacity for low-temperature refrigeration applications, the vapor-injection strategy is innovatively synthesized with two-stage cascade refrigeration systems. Two cascade vapor-injection configurations with subcoolers and flash tanks (CSVIRS and CFVIRS) are compared with the conventional cascade refrigeration system (CCRS) through integrated thermodynamic simulations. The impacts of crucial temperature and injection parameters are comprehensively analyzed through energy and exergy methods, while the performance comparisons of various refrigerant combinations are also conducted. The coefficient of performance (COP) of the CFVIRS exceeds that of the CCRS and CSVIRS by 33.84% and 2.10% under the default condition. The cascade vapor-injection configurations exhibit a performance advantage at higher condensation temperature and lower evaporation temperature of the low-temperature cycle (LTC). The evaporation temperature of the high-temperature cycle (HTC) and injection pressures are examined with optimum solutions. Decreasing the entrainment ratio of the HTC and increasing the entrainment ratio of the LTC within appropriate ranges are beneficial for the refrigeration performance. R1270-R170 demonstrates superior energy and exergy performance, whereas R143a-R23 shows the highest improvement ratio among the compared refrigerants. The implementation of cascade vapor injection substantially reduces exergy destruction in the compression and expansion devices, while the exergy characteristics of various refrigerant pairs are extensively investigated.

This study investigates the unsteady gas–liquid two-phase flow evolution and associated thermal transport mechanisms driven by a moving interface within confined spaces. Utilizing Large Eddy Simulation coupled with the Volume of Fluid method, we characterize the complete dynamical pathway—transitioning from initial axisymmetric ordered convection to localized symmetry breaking, and ultimately to global chaotic turbulence. Particular emphasis is placed on the regulatory role of geometric constraints (aspect ratio λ) on flow topology and instability modes. Physical analysis reveals that high-aspect-ratio configurations suppress the nonlinear growth of radial perturbations by enhancing the coupling between sidewall confinement and viscous dissipation, thereby effectively retarding the spatiotemporal diffusion of turbulence. Regarding thermal transport, the breakdown of ordered convective structures is closely associated with the onset of nonlinear growth in gas temperature. Based on dimensionless analysis, we clarify the competitive mechanism among the timescales of perturbation growth, thermal diffusion, and characteristic compression. The results demonstrate that strong geometric constraints establish a near-isothermal physical limit by maintaining macroscopic ordered flow organization. This work provides new physical insights into the coupled “flow regime-thermal transport” laws in confined unsteady flows, offering a theoretical foundation for understanding complex interfacial dynamics.

Liquid air energy storage (LAES) systems utilize off-peak or renewable electricity to produce liquid air, which serves as a promising storage medium featuring high energy density, environmental friendliness and operational flexibility. Conventional coupled LAES systems rely on integration with industrial processes or external energy sources. However, the spatiotemporal mismatches between energy storage and electricity consumption necessitate advanced decoupled LAES configurations capable of efficient heat compensation and cryogenic energy recovery. To address this challenge, a novel decoupled semi-closed LAES system is proposed for micro-grid applications. The system integrates two vapor-injection heat pumps (VIHPs) and a vapor-injection refrigeration system (VIRS) to extract heat from ambient sources and recover cryogenic energy for air re-liquefaction, respectively. Comprehensive thermodynamic analyses and parameter optimizations are conducted to validate the performance improvement over a conventional decoupled LAES system. The results show that the integration of VIHPs and VIRS significantly enhances energy efficiency and reduces liquid air consumption. The proposed system achieves the optimal round-trip efficiency, combined cooling and power efficiency and specific liquid consumption of 28.18%, 54.35% and 7.96 kg/kWh. R1234ze(Z) is recommended as the optimal high-temperature refrigerant for heat supplementation, while replacing purified air with nitrogen as the storage medium can further improve the round-trip efficiency by 4.21%.

Static mixers play a vital role in ensuring the uniform mixing of natural gas (NG) and hydrogen, thereby reducing operational costs at gas transmission stations and enhancing transportation efficiency. This study, based on the design parameters of a section of the high-pressure Shaanxi-Beijing Gas Pipeline I, employs numerical simulations to compare the flow characteristics and mixing mechanisms of four types of static mixers: T-junction, Kenics, High-Efficiency Vane (HEV), and Coaxial shear. A hydrogen blending experimental platform was constructed, and experimental investigations were carried out to validate the strong agreement between numerical simulations and experimental data. The study explores the velocity distribution patterns and mixing behaviors of the different mixers and analyzes the effects of hydrogen blending ratios - ranging from 5% to 20% - on mixing uniformity and pressure drop. The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is adopted to provide a comprehensive performance evaluation and ranking of the four mixers. The results indicate that the performance ranking from highest to lowest is: Coaxial shear mixer (Score: 0.84, Rank: 1), Kenics mixer (Score: 0.59, Rank: 2), HEV mixer (Score: 0.53, Rank: 3), and T-junction mixer (Score: 0.40, Rank: 4). Under identical operating conditions, the Coaxial shear mixer achieved the shortest distance required for complete uniform mixing (1D) and a relatively low pressure drop (5321.12 Pa), while the T-junction mixer required the longest mixing distance (43D) but had the lowest pressure drop (3033.99 Pa). This study provides experimental data for hydrogen-blended static mixers and offers methodological guidance for selecting static mixers in industrial applications.

Hydrogen-blended natural gas delivered via pipeline networks supports the development of hydrogen energy. However, hydrogen-blended natural gas containing water vapour significantly decreases delivery efficiency and accelerates pipeline corrosion, necessitating rigorous dewatering before transportation, storage, and utilisation. This mixture is primarily methane-based, with component variations depending on the source. This study employs Box–Behnken design (BBD) and particle swarm optimisation (PSO) to examine how the composition and content of hydrogen-blended natural gas influence the condensation rate in Laval nozzles. Utilising computational fluid dynamics (CFD), we calculated the flow parameters of various hydrogen-blended natural gas fractions and contents through the Laval nozzle. The optimal gas composition was identified using BBD and PSO, resulting in a condensation rate of 78.65% with CH4 at 69.212%, C2H6 at 2.028%, C3H8 at 1.107%, and H2 at 9.01%. This condensation rate is significantly higher compared to mixtures containing only CH4 and H2. These findings guide the investigation of phase transformation mechanisms in multi-component condensation of hydrogen-blended natural gas with water vapour and their engineering applications.

Hydrogen energy is widely recognized as the clean energy source with the most potential. Blending hydrogen into natural gas pipelines is as effective approach to address the problem of hydrogen transport. The exploitation of natural gas and hydrogen production involve water vapour. Therefore, when hydrogen-blended natural gas (HBNG) containing water vapour flows through a complex natural gas pipeline network, the water vapour can condense, which can cause the pipeline network to malfunction. Therefore, in the present work, a condensation model of water vapour-containing HBNG in a Laval nozzle was calculated by computational fluid dynamics method; And the convergence section adopted four curves: bicubic, quintic polynomial, Witoszynski, and Witoszynski translation equations. The calculations showed that the Witoszynski equation curve achieved excellent condensation performance, with a maximum liquid mass fraction of 0.0277. Furthermore, the effects of structural parameters of the Witoszynski equation curve, composition of HBNG components comprising water vapour, inlet temperature, and pressure on condensation characteristics were investigated. With increasing shrinkage ratio, divergent angle, and divergent length, the maximum liquid mass fraction increased by 4.69%, 9.79%, and 0.35%, respectively. When the water vapour content and inlet pressure increased gradually, the maximum liquid mass fraction increased by 1.06% and 40.98%, respectively. With increasing hydrogen blending ratio and inlet temperature, the maximum liquid mass fraction decreased by 7.48% and 15.53%, respectively. The findings of this study offer a valuable reference for optimizing the design and regulating the condensation performance of water vapour-containing HBNG within Laval nozzles.

ABSTRACT Magnetic flux leakage detection is a well‐established non‐destructive in‐line inspection for pipelines. In practical applications, the volume of inspection data is large, and manually labeling of defects is time‐consuming and inefficient. Automated processing of inspection data using machine learning methods can address these issues. This study evaluates the performance of four machine learning models in defect classification and defect depth prediction. The synthetic minority oversampling technique (SMOTE) is employed to increase the number of minority class samples, enhancing the model's generalization ability and thereby improving the classification accuracy for minority class defects. The prediction results showed that the Categorical Boosting (CatBoost) model had a defect classification accuracy of 0.9730, a mean squared error of 0.0256 for defect depth prediction, and a prediction residual range of −1 to 1.1 mm. The CatBoost model had the advantages of high classification accuracy, a small prediction error, and a residual range. The study results provided a reference for predicting pipeline defect types and defect depths using machine learning models.

In long-distance oil and gas pipelines, defects are generated owing to the extension of service life and environmental impacts. In practical for in-line inspection(ILI), several challenges exist, including long pipeline mileage, large volume of inspection data, numerous abnormal signals, and low efficiency of manual defect identification. To achieve efficient processing of inspection data and automatic identification of pipeline defects, a complete pipeline defect inspection process and the Multi-Branch Depth-Sense Fusion Network (MBDSF-Net) have been proposed. For the obtained ILI data, in the stage of inspection data analysis and processing, mileage accuracy optimisation and abnormal signal elimination operations are implemented. Additionally, an efficient method for plotting inspection data is proposed. This method generates inspection images for 316.87 km within 1 h, achieving rapid conversion of inspection data into images. The detection network utilises Inception Depthwise Convolution (IDWC), which expands the receptive field while improving computational efficiency and model accuracy. Furthermore, the Multi-Scale Deep Feature Fusion Network (MSDFF-Net) is proposed to enhance the representation capabilities of features at each level. The Inception-enhanced Mixed Aggregation Network (IMANet) is developed to reinforce backbone feature extraction to achieve more discriminative representations. Experimental results demonstrate that MBDSF-Net achieves an mAP50 of 91.0% and an inference latency of 4.01 ms on the self-built PIP-813 dataset, achieving a desirable balance between accuracy and speed. The proposed inspection data processing method and MBDSF-Net improve the processing efficiency of inspection data and the accuracy of pipeline defect identification, making them highly suitable for practical engineering applications.

The depth of metal loss defects is crucial for assessing the safety of pipelines. Magnetic flux leakage (MLF) testing is an efficient and non-destructive testing technique, but the sensor output is magnetic flux leakage data, which cannot directly provide defect size information. In actual pipeline inspection, the amount of data obtained by the sensor is huge, and manual defect annotation is time-consuming and labor-intensive. To understand the safety status of pipelines and save labor costs, artificial intelligence methods need to be used to process and analyze the magnetic flux leakage data to determine whether defects exist and predict the depth of defects. [Method] This paper uses random forest regression, gradient boosting decision tree, support vector regression, and convolutional neural network methods to analyze the magnetic flux leakage detection data and compare their performance in defect depth prediction. [Result] The experimental results show that support vector regression has the lowest requirement for computing power but has a relatively large prediction error. The convolutional neural network has the smallest prediction error and the smallest residual range, but it has a high requirement for computing power. The gradient boosting decision tree has the smallest average error and a low requirement for computing power. [Conclusion] Considering the computational cost and prediction accuracy, the gradient boosting decision tree is the best choice, with a mean absolute error (MAE) of 0.2329, which is smaller than that of previous studies. The method proposed in this paper can serve as a reference for predicting defect size using magnetic flux leakage data.

This study investigates the blending characteristics of natural gas (NG) and hydrogen in a Kenics static mixer using computational fluid dynamics. The effectiveness of the adopted numerical model is experimentally validated. The scenario of high-pressure, long-distance NG pipelines is considered, and the Soave–Redlich–Kwong equation of state is applied. The mixing uniformity and pressure loss are adopted as evaluation criteria to analyze the impact of factors such as the deflection angle of spiral blades, the number of blades, the length-to-diameter ratio, the hydrogen blending ratio (HBR), the pipeline pressure, and the temperature on mixer performance, followed by structural optimization of the mixer. It is found that a Kenics static mixer with two spiral blades, a length-to-diameter ratio of 2, and a deflection angle of 135° can achieve a mixing uniformity of 95% at the blade outlet while minimizing pressure loss. Increasing the HBR helps improve the mixing uniformity but also increases the pressure loss of the mixer. Increasing the pipeline pressure while keeping the hydrogen mole fraction constant enhances the mixing uniformity but also increases the pressure loss. Increasing the gas temperature reduces the mixing uniformity and the pressure loss. Overall, pipeline pressure and temperature changes have a minimal impact on the mixing characteristics. Under high-pressure conditions, the use of a real gas model is essential. This study provides theoretical guidance for the design of static mixers for hydrogen blending in high-pressure NG pipelines.

Hydrogen-blended natural gas (NG) pipeline network transport is the most effective approach for solving the problem of large-scale hydrogen use. Hydrogen-blended NG that contains water vapour is prone to water vapour condensation when it passes through complex NG pipeline networks, leading to pipeline network failures. To analyse the condensation behaviour of hydrogen-blended NG containing water vapour in a Laval nozzle, a condensation model of water vapour was established. A computational fluid dynamics approach was used to calculate the condensation process of hydrogen-blended NG containing water vapour in Laval nozzles for four countries: Iran, USA, Russia, and Australia. Hydrogen-blended NG components affect the flow characteristics of the gas mixture in the nozzle. The gas components have the greatest effect on the Mach number. The difference between the maximum and minimum Mach numbers at the outlet was 0.02 Mach. Hydrogen-blended NG containing water vapour condenses downstream of the throat of the Laval nozzle. Hydrogen-blended NG from Russia had the largest condensation ratio (79.63 %). The largest droplet radius and liquid mass fraction were observed in the hydrogen-blended NG from Australia. The condensation process can accelerate the future research and engineering application of water vapour into hydrogen-blended NG.

In order to solve the problem of thickness measurement of stainless steel metal plate in high-temperature application, the electromagnetic ultrasonic technology is greatly promising due to its non-touch feature. In this paper, an alternating double coils for excitation and reception is designed, and the propagation characteristics of electromagnetic ultrasonic wave in stainless steel are researched by the COMSOL software. The simulation results show that the eddy currents induced on the surface of the specimen can form shear waves propagating downward, and the internal stress reached 4.75 E 5 Pa. In addition, the electromagnetic ultrasonic thickness measurement system was designed and built. Using pulse compression technology, the Signal-to-Noise Ratio (SNR) of echo signals have been greatly improved. The thickness of stainless steel metal specimens under different temperatures were tested, and the measurement error was analyzed in detail. The results of the experiments indicate that the system has an excellent measurement performance for metal thickness. At a high temperature of $600^{\circ} \mathrm{C}$, the SNR can reach 22 dB, and the thickness error is within $5 \%$.

Hydrogen delivery through existing natural gas pipeline infrastructure offers a sustainable and cost-effective approach for large-scale distribution while maintaining operational efficiency. In such systems, the high-pressure regulator valves are critical to control the hydrogen flow. However, because of the physical differences between hydrogen and natural gas, hydrogen aggregation can occur during blended natural gas transport. This study employs computational fluid dynamics to investigate how valve opening, hydrogen blending ratio, inlet velocity, temperature, and pressure influence hydrogen concentration distribution within the valve. The results demonstrate that smaller valve openings lead to steeper hydrogen concentration gradients along the pipeline, with the gradient at a 5% valve opening reaching four times that observed at 20%. While increased inlet velocity mitigates these gradients, high flow rates introduce uneven hydrogen distributions near the valve. Higher hydrogen blending ratios and operating pressures further exacerbate concentration non-uniformity. These findings offer critical insights for enhancing safety and performance in hydrogen-blended natural gas delivery systems.

Hydrogen-blended natural gas containing water vapour is susceptible to condensation in the regulating valve with risks of clogging and corrosion. However, there is a lack of targeted research and analysis on the condensation characteristics of the regulating valve. A regulating-valve model is constructed to analyse the pressure, temperature distribution, and condensation process of hydrogen-blended natural gas containing water vapour using the computational fluid dynamics method. The arc cone-valve core regulating valves have a larger flow area than the straight and flat-bottomed cone-valve cores for the same opening and also have a lower pressure decrease and condensation rate. The condensation rate is reduced by 18.31 % when the hydrogen blending volume ratio is increased from 5 % to 90 %. The smaller the opening, the greater the condensation rate in regulating valve. With an increasing opening, the magnitudes of the decreases in temperature, pressure, condensation rate after the flow of hydrogen-blended natural gas through the valve core were greater at openings smaller than 30 %. The magnitudes of these changes decrease when the opening is greater than 30 %. The study complements the lack of research on the flow and condensation characteristics and provides a reference for the safe operation in hydrogen-blended natural gas projects.