Catalytic combustion synthesis of CNTs/MgO composite powders and the influences on the thermal shock resistance of low-carbon Al2O3–C refractories

Using MgC₂O₄, Mg powders as raw materials and Ni(NO₃)₂?6H₂O as a catalyst, CNTs/MgO composite powders were prepared by a catalytic combustion synthesis method. The CNTs/MgO composite powders were characterized by XRD, Raman spectroscopy, FESEM/EDS and HRTEM. The effects of catalyst content on the degree of graphitization and aspect ratio of the CNTs in composite powders were investigated. Moreover, the thermal shock resistance of low-carbon Al₂O₃–C refractories after adding the composite powder was investigated. The results indicated that the CNTs prepared with 1 wt% Ni(NO₃)₂?6H₂O addition had a higher degree of graphitization and aspect ratio. In particular, the aspect ratio could reach approximately 200. The growth mechanism of hollow bamboo-like CNTs in the composite powders was proven to be a V-L-S mechanism. The thermal shock resistance of Al₂O₃–C samples could be improved significantly after adding CNTs/MgO composite powders. In particular, compared with CM0, the residual strength ratio of Al₂O₃–C samples with added 2.5 wt% composite powders could be increased 63.9%.     

Tabular data: Deep learning is not all you need

A key element in solving real-life data science problems is selecting the types of models to use. Tree ensemble models (such as XGBoost) are usually recommended for classification and regression problems with tabular data. However, several deep learning models for tabular data have recently been proposed, claiming to outperform XGBoost for some use cases. This paper explores whether these deep models should be a recommended option for tabular data by rigorously comparing the new deep models to XGBoost on various datasets. In addition to systematically comparing their performance, we consider the tuning and computation they require. Our study shows that XGBoost outperforms these deep models across the datasets, including the datasets used in the papers that proposed the deep models. We also demonstrate that XGBoost requires much less tuning. On the positive side, we show that an ensemble of deep models and XGBoost performs better on these datasets than XGBoost alone.     

Achieving superior hydrogen storage properties via in-situ formed nanostructures: A high-capacity Mg–Ni alloy with La microalloying

Mg-based hydride is a promising hydrogen storage material, but its capacity is hindered by the kinetic properties. In this study, Mg–Mg₂Ni–LaH_x nanocomposite is formed from the H-induced decomposition of Mg₉₈Ni_1·67La_0.33 alloy. The hydrogen capacity of 7.19 wt % is reached at 325 °C under 3 MPa H₂, attributed to the ultrahigh hydrogenation capacity in Stage I. The hydrogen capacity of 5.59 wt % is achieved at 175 °C under 1 MPa H₂. The apparent activation energies for hydrogen absorption and desorption are calculated as 57.99 and 107.26 kJ/mol, which are owing to the modified microstructure with LaH_x and Mg₂Ni nanophases embedding in eutectic, and tubular nanostructure adjacent to eutectic. The LaH_2.49 nanophase can catalyze H₂ molecules to dissociate and H atoms to permeate due to its stronger affinity with H atoms. The interfaces of these nanophases provide preferential nucleation sites and alleviate the “blocking effect” together with tubular nanostructure by providing H atoms diffusion paths after the impingement of MgH₂ colonies. Therefore, the superior hydrogenation properties are achieved because of the rapid absorption process of Stage I. The efficient synthesis of nano-catalysts and corresponding mechanisms for improving hydrogen storage properties have important reference to related researches.     

Data fusion and transfer learning empowered granular trust evaluation for Internet of Things

In the Internet of Things (IoT), a huge amount of valuable data is generated by various IoT applications. As the IoT technologies become more complex, the attack methods are more diversified and can cause serious damages. Thus, establishing a secure IoT network based on user trust evaluation to defend against security threats and ensure the reliability of data source of collected data have become urgent issues, in this paper, a Data Fusion and transfer learning empowered granular Trust Evaluation mechanism (DFTE) is proposed to address the above challenges. Specifically, to meet the granularity demands of trust evaluation, time–space empowered fine/coarse grained trust evaluation models are built utilizing deep transfer learning algorithms based on data fusion. Moreover, to prevent privacy leakage and task sabotage, a dynamic reward and punishment mechanism is developed to encourage honest users by dynamically adjusting the scale of reward or punishment and accurately evaluating users’ trusts. The extensive experiments show that: (i) the proposed DFTE achieves high accuracy of trust evaluation under different granular demands through efficient data fusion; (ii) DFTE performs excellently in participation rate and data reliability.     

Learn decision trees with deep visual primitives

In this paper, we strive to propose a self-interpretable framework, termed PrimitiveTree, that incorporates deep visual primitives condensed from deep features with a conventional decision tree, bridging the gap between deep features extracted from deep neural networks (DNNs) and trees’ transparent decision-making processes. Specifically, we utilize a codebook, which embeds the continuous deep features into a finite discrete space (deep visual primitives) to distill the most common semantic information. The decision tree adopts the spatial location information and the mapped primitives to present the decision-making process of the deep features in a tree hierarchy. Moreover, the trained interpretable PrimitiveTree can inversely explain the constituents of the deep features, highlighting the most critical and semantic-rich image patches attributing to the final predictions of the given DNN. Extensive experiments and visualization results validate the effectiveness and interpretability of our method.     

A data-driven digital-twin prognostics method for proton exchange membrane fuel cell remaining useful life prediction

Prognostics and health management of proton exchange membrane fuel cell (PEMFC) systems have driven increasing research attention in recent years as the durability of PEMFC stack remains as a technical barrier for its large-scale commercialization. To monitor the health state during PEMFC operation, digital twin (DT), as a smart manufacturing technique, is applied in this paper to establish an ensemble remaining useful life prediction system. A data-driven DT is constructed to integrate the physical knowledge of the system and a deep transfer learning model based on stacked denoising autoencoder is used to update the DT with online measurement. A case study with experimental PEMFC degradation data is presented where the proposed data-driven DT prognostics method has applied and reached a high prediction accuracy. Furthermore, the predicted results are proved to be less affected even with limited measurement data.     

高深溜井井筒堵塞的爆破处理实践

溜井是采用平硐-溜井方式开拓矿山的运输咽喉,溜井正常与否对矿山生产影响极大。本文通过黑沟铁矿高深溜井井筒堵塞处理实例,对其堵塞爆破处理方法及经验做了系统分析。重点介绍的爆破冲击波破拱疏通高深溜井井筒高位堵塞的爆破方法,富有新意,可供国内外同类型矿山参考借鉴。     

Preparation of Pt/Ce–Zr–Y mixed oxide/anodized aluminium oxide catalysts for hydrogen passive autocatalytic recombination

The use of a Pt-based catalyst was evaluated for autocatalytic hydrogen recombination. The Pt was supported on a mixture of Ce-, Zr- and Y-oxides (CZY) to yield nanosized Pt particles. The Pt/CZY/AAO catalyst was then prepared by the spray-deposition of the Pt/CZY intermediate onto an anodized aluminium oxide (AAO) layer on a metallic aluminum core. The Pt/CZY/AAO catalyst (3 × 1 cm) was evaluated for hydrogen combustion (1–8 vol% hydrogen in the air) in a recombiner section testing station. The thermal distribution throughout the catalyst surface was investigated using an infrared camera. The maximum temperature gradient (ΔT) for the examined hydrogen concentrations did not exceed 36 °C. The Pt/CZY/AAO catalyst was also evaluated for prolonged hydrogen combustion duration to assess its durability. An average combustion temperature of 239.0 ± 10.0 °C was maintained for 53 days of catalytic hydrogen combustion, suggesting that there was limited, or no, catalyst deactivation. Finally, a Pt/CZY/AAO catalytic plate (14.0 × 4.5 cm) was prepared to investigate the thermal distribution. An average surface temperature of 212.5 °C and a maximum ΔT of 5.4 °C was obtained throughout the catalyst surface at a 3 vol% hydrogen concentration.     

Highly stable nanocarbon supported Pt catalyst for fuel cell via a molten salt graphitization strategy

Proton exchange membrane fuel cells (PEMFCs) durability has been severely hindered by carbon support poor stability in the cathodic Pt-based catalyst. Herein, a high-surface-area nitrogen-doped graphitic nanocarbon (N-G-CA) with mesopores is developed as Pt support to address PEMFCs durability challenge. Resorcinol-formaldehyde aerogel pyrolyzed carbon aerogel is selected as N-G-CA raw material. Nitrogen atoms are introduced into carbon aerogel via NH₃ heat treatment. Then, nitrogen-doped carbon aerogel is transferred into N-G-CA via heating together with transition-metal salts (one of FeCl₃, FeCl₂, CoCl₂, or MnCl₂, etc.) at 1200 °C. As ORR catalyst, Pt/N-G-CA half-wave potential only lost 10 mV, after 30, 000 cycles accelerated aging test in the rotating-desk-electrode. Only 12 mV voltage loss at 1.5 A/cm2 is observed, after 5, 000 cycles for membrane electrode. Pt/N-G-CA exhibits superior durability and activity than commercial Pt/C. High durability of Pt/N-G-CA is due to N-G-CA high graphitization extent, as well as the interactions between doping nitrogen and Pt. N-G-CA is promising as stable support for durable Pt-based catalysts in PEMFCs, thanks to enhanced carbon corrosion resistance, uniformly dispersed Pt, and strong support-metals interaction.     

Salt-assisted solution combustion synthesis of NiFe2O4: Effect of salt type

The effects of three types of salt including NaF, KCl, and NaCl on the properties of NiFe₂O₄ nanoparticles using salt-assisted solution combustion synthesis (SSCS) have been investigated. The synthesized powders were evaluated by SEM, TEM, FTIR, XRD, and VSM analysis. Also, the specific surface area (SSA), as well as size distribution and volume of the porosities of NiFe₂O₄ powders were determined by the BET apparatus. The visual observations showed that the intensity and time of combustion synthesis of nanoparticles have been severely influenced by the type of salt. The highest crystallinity was observed in the synthesized powder using NaCl. The SSA has also been correlated completely to the type of salt. The quantities of SSA was achieved about 91.62, 64.88, and 47.22 m²g^-1 for the powders synthesized by KCl, NaCl, and NaF respectively. Although the magnetic hysteresis loops showed the soft ferromagnetic behavior of the NiFe₂O₄ nanoparticles in all conditions, KCl salt could produce the particles with the least coercivity and remanent magnetization. Based on the present study, the salt type is a key parameter in the SSCS process for the preparation of spinel ferrites. Thermodynamic evaluation also showed that the melting point and heat capacity are important parameters for the proper selection of the salt.