Changsha University of Science and Technology

This paper employs the Auto-Encoding Variational Bayes (AEVB) estimator based on Stochastic Gradient Variational Bayes (SGVB), designed to optimize recognition models for challenging posterior distributions and large-scale datasets. It has been applied to the mnist dataset and extended to form a Dynamic Bayesian Network (DBN) in the context of time series. The paper delves into Bayesian inference, variational methods, and the fusion of Variational Autoencoders (VAEs) and variational techniques. Emphasis is placed on reparameterization for achieving efficient optimization. AEVB employs VAEs as an approximation for intricate posterior distributions.

The reasonable design of electrode materials for rechargeable batteries plays an important role in promoting the development of renewable energy technology. With the in-depth understanding of the mechanisms underlying electrode reactions and the rapid development of advanced technology, the performance of batteries has significantly been optimized through the introduction of defect engineering on electrode materials. A large number of coordination unsaturated sites can be exposed by defect construction in electrode materials, which play a crucial role in electrochemical reactions. Herein, recent advances regarding defect engineering in electrode materials for rechargeable batteries are systematically summarized, with a special focus on the application of metal-ion batteries, lithium-sulfur batteries, and metal-air batteries. The defects can not only effectively promote ion diffusion and charge transfer but also provide more storage/adsorption/active sites for guest ions and intermediate species, thus improving the performance of batteries. Moreover, the existing challenges and future development prospects are forecast, and the electrode materials are further optimized through defect engineering to promote the development of the battery industry.

Abstract Demand for energy in day to day life is increasing exponentially. However, existing energy storage technologies like lithium ion batteries cannot stand alone to fulfill future needs. In this regard, potassium ion batteries (KIBs) that utilize K ions in their charge storage mechanism have attracted considerable attention due to their unique properties and are therefore established as one of the future battery systems of interest among the scientific community. Nevertheless, the development and identification of appropriate electrode materials is very essential for practical applications. This review features the current development in KIBs electrode and electrolyte materials, the present challenges facing this technology (in the commercial aspect), and future aspects to develop fully functional KIBs. The potassium storage mechanisms, evolution of the KIBs, and the advantages and disadvantages of each category of materials are included. Additionally, various approaches to enhance the electrochemical performances of KIBs are also discussed. This review is not only an amalgamation of different viewpoints in literature, but also contains concise perspectives and strategies. Moreover, the potential emergence of a novel class of K‐based dual ion batteries is also analyzed for the first time.

Single-layer chromium trihalides constitute a series of stable 2D intrinsic FM half semiconductors with large magnetic anisotropy energies.

1,2,3,5-Tetrakis(carbazol-9-yl)-4,6-dicyanobenzene (4CzIPN) is a typical donor-acceptor fluorophore, with carbazolyl as an electron donor and dicyanobenzene as an electron acceptor. It has emerged as a powerful organophotocatalyst since 2016. Excellent redox window, good chemical stability and broad applicability make 4CzIPN an attractive metal-free photocatalyst. In this review, the recent advances of the application of 4CzIPN as a photoredox catalyst in the past three years (2016-2018) for various organic reactions are summarized.

Abstract Iron phthalocyanine (FePc) is a promising non-precious catalyst for the oxygen reduction reaction (ORR). Unfortunately, FePc with plane-symmetric FeN 4 site usually exhibits an unsatisfactory ORR activity due to its poor O 2 adsorption and activation. Here, we report an axial Fe–O coordination induced electronic localization strategy to improve its O 2 adsorption, activation and thus the ORR performance. Theoretical calculations indicate that the Fe–O coordination evokes the electronic localization among the axial direction of O–FeN 4 sites to enhance O 2 adsorption and activation. To realize this speculation, FePc is coordinated with an oxidized carbon. Synchrotron X-ray absorption and Mössbauer spectra validate Fe–O coordination between FePc and carbon. The obtained catalyst exhibits fast kinetics for O 2 adsorption and activation with an ultralow Tafel slope of 27.5 mV dec −1 and a remarkable half-wave potential of 0.90 V. This work offers a new strategy to regulate catalytic sites for better performance.

Abstract A critical challenge in the commercialization of layer‐structured Ni‐rich materials is the fast capacity drop and voltage fading due to the interfacial instability and bulk structural degradation of the cathodes during battery operation. Herein, with the guidance of theoretical calculations of migration energy difference between La and Ti from the surface to the inside of LiNi 0.8 Co 0.1 Mn 0.1 O 2 , for the first time, Ti‐doped and La 4 NiLiO 8 ‐coated LiNi 0.8 Co 0.1 Mn 0.1 O 2 cathodes are rationally designed and prepared, via a simple and convenient dual‐modification strategy of synchronous synthesis and in situ modification. Impressively, the dual modified materials show remarkably improved electrochemical performance and largely suppressed voltage fading, even under exertive operational conditions at elevated temperature and under extended cutoff voltage. Further studies reveal that the nanoscale structural degradation on material surfaces and the appearance of intergranular cracks associated with the inconsistent evolution of structural degradation at the particle level can be effectively suppressed by the synergetic effect of the conductive La 4 NiLiO 8 coating layer and the strong TiO bond. The present work demonstrates that our strategy can simultaneously address the two issues with respect to interfacial instability and bulk structural degradation, and it represents a significant progress in the development of advanced cathode materials for high‐performance lithium‐ion batteries.

The commercialization of fuel cells, such as proton exchange membrane fuel cells and direct methanol/formic acid fuel cells, is hampered by their poor stability, high cost, fuel crossover, and the sluggish kinetics of platinum (Pt) and Pt-based electrocatalysts for both the cathodic oxygen reduction reaction (ORR) and the anodic hydrogen oxidation reaction (HOR) or small molecule oxidation reaction (SMOR). Thus far, the exploitation of active and stable electrocatalysts has been the most promising strategy to improve the performance of fuel cells. Accordingly, increasing attention is being devoted to modulating the surface/interface electronic structure of electrocatalysts and optimizing the adsorption energy of intermediate species by defect engineering to enhance their catalytic performance. Defect engineering is introduced in terms of defect definition, classification, characterization, construction, and understanding. Subsequently, the latest advances in defective electrocatalysts for ORR and HOR/SMOR in fuel cells are scientifically and systematically summarized. Furthermore, the structure-activity relationships between defect engineering and electrocatalytic ability are further illustrated by coupling experimental results and theoretical calculations. With a deeper understanding of these complex relationships, the integration of defective electrocatalysts into single fuel-cell systems is also discussed. Finally, the potential challenges and prospects of defective electrocatalysts are further proposed, covering controllable preparation, in situ characterization, and commercial applications.

In recent years, with the rapid development of Internet technology, online shopping has become a mainstream way for users to purchase and consume. Sentiment analysis of a large number of user reviews on e-commerce platforms can effectively improve user satisfaction. This paper proposes a new sentiment analysis model-SLCABG, which is based on the sentiment lexicon and combines Convolutional Neural Network (CNN) and attention-based Bidirectional Gated Recurrent Unit (BiGRU). In terms of methods, the SLCABG model combines the advantages of sentiment lexicon and deep learning technology, and overcomes the shortcomings of existing sentiment analysis model of product reviews. The SLCABG model combines the advantages of the sentiment lexicon and deep learning techniques. First, the sentiment lexicon is used to enhance the sentiment features in the reviews. Then the CNN and the Gated Recurrent Unit (GRU) network are used to extract the main sentiment features and context features in the reviews and use the attention mechanism to weight. And finally classify the weighted sentiment features. In terms of data, this paper crawls and cleans the real book evaluation of dangdang.com, a famous Chinese e-commerce website, for training and testing, all of which are based on Chinese. The scale of the data has reached 100000 orders of magnitude, which can be widely used in the field of Chinese sentiment analysis. The experimental results show that the model can effectively improve the performance of text sentiment analysis.

The relational facts in sentences are often complicated. Different relational triplets may have overlaps in a sentence. We divided the sentences into three types according to triplet overlap degree, including Normal, EntityPairOverlap and SingleEn-tiyOverlap. Existing methods mainly focus on Normal class and fail to extract relational triplets precisely. In this paper, we propose an end-to-end model based on sequence-to-sequence learning with copy mechanism, which can jointly extract relational facts from sentences of any of these classes. We adopt two different strategies in decoding process: employing only one united decoder or applying multiple separated decoders. We test our models in two public datasets and our model outperform the baseline method significantly.

Mixed convection is a mechanism of heat transport in a thermodynamic system in which the motion of fluid particles is produced by gravity as well as external forces like fans, pumps, or any other devices. Such type of heat transport has a fruitful application in daily life due to reliable maintenance. In this regard, numerous researchers and analyst have focused on the importance of mixed convective flow to explore its different aspects, and frequent research articles are published in this area. In this work, mixed convective entropy optimized nanomaterial magnetohydrodynamics (MHD) flow of Ree‐Eyring fluid is discussed between two rotating disks. The effects of porosity and velocity slip are considered. Both the disks are rotating with different angular frequency and stretching rates. Modeling is performed for the energy equation subject to heat generation/absorption, dissipation, radiative heat flux, and Joule heating. Four types of irreversibilities are discussed, and total entropy rate is calculated. The obtained results are compared with past studies and found good agreement with them. The physical curiosity like skin friction and Sherwood and Nusselt numbers are numerically calculated. Series solutions are computed via homotopy method. Our obtained outcomes show that the velocity and temperature fields show contrast behavior against larger magnetic parameter. It is also noticed that the entropy rate and Bejan number have opposite behaviors against higher values of Weissenberg number. The entropy rate increases for higher Weissenberg number while Bejan number decays.

Laser powder bed fusion (LPBF) has made significant progress in producing solid and porous metal parts with complex shapes and geometries. However, LPBF produced parts often have defects (e.g., porosity, residual stress, and incomplete melting) that hinder its large-scale industrial commercialization. The LPBF process involves complex heat transfer and fluid flow, and the melt pool is a critical component of the process. The melt pool stability is a critical factor in determining the microstructure, mechanical properties, and corrosion resistance of LPBF produced metal parts. Furthermore, optimizing process parameters for new materials and designed structures is challenging due to the complexity of the LPBF process. This requires numerous trial-and-error cycles to minimize defects and enhance properties. This review examines the behavior of the melt pool during the LPBF process, including its effects and formation mechanisms. This article summarizes the experimental results and simulations of melt pool and identifies various factors that influence its behavior, which facilitates a better understanding of the melt pool's behavior during LPBF. This review aims to highlight key aspects of the investigation of melt pool tracks and microstructural characterization, with the goal of enhancing a better understanding of the relationship between alloy powder-process-microstructure-properties in LPBF from both single- and multi-melt pool track perspectives. By identifying the challenges and opportunities in investigating single- and multi-melt pool tracks, this review could contribute to the advancement of LPBF processes, optimal process window, and quality optimization, which ultimately improves accuracy in process parameters and efficiency in qualifying alloy powders.

Energy management is an enabling technology for increasing the economy of fuel cell/battery hybrid electric vehicles. Existing efforts mostly focus on optimization of a certain control objective (e.g., hydrogen consumption), without sufficiently considering the implications for on-board power sources degradation. To address this deficiency, this article proposes a cost-optimal, predictive energy management strategy, with an explicit consciousness of degradation of both fuel cell and battery systems. Specifically, we contribute two main points to the relevant literature, with the purpose of distinguishing our study from existing ones. First, a model predictive control framework, for the first time, is established to minimize the total running cost of a fuel cell/battery hybrid electric bus, inclusive of hydrogen cost and costs caused by fuel cell and battery degradation. The efficacy of this framework is evaluated, accounting for various sizes of prediction horizon and prediction uncertainties. Second, the effects of driving and pricing scenarios on the optimized vehicular economy are explored.

This review paper summarizes the categories, sensing mechanisms, and affecting factors of flexible conductive polymer composite-based stretchable strain sensors.

Internet of Things (IoT) realizes the interconnection of heterogeneous devices by the technology of wireless and mobile communication. The data of target regions are collected by widely distributed sensing devices and transmitted to the processing center for aggregation and analysis as the basis of IoT. The quality of IoT services usually depends on the accuracy and integrity of data. However, due to the adverse environment or device defects, the collected data will be anomalous. Therefore, the effective method of anomaly detection is the crucial issue for guaranteeing service quality. Deep learning is one of the most concerned technology in recent years which realizes automatic feature extraction from raw data. In this article, the integrated model of the convolutional neural network (CNN) and recurrent autoencoder is proposed for anomaly detection. Simple combination of CNN and autoencoder cannot improve classification performance, especially, for time series. Therefore, we utilize the two-stage sliding window in data preprocessing to learn better representations. Based on the characteristics of the Yahoo Webscope S5 dataset, raw time series with anomalous points are extended to fixed-length sequences with normal or anomaly label via the first-stage sliding window. Then, each sequence is transformed into continuous time-dependent subsequences by another smaller sliding window. The preprocessing of the two-stage sliding window can be considered as low-level temporal feature extraction, and we empirically prove that the preprocessing of the two-stage sliding window will be useful for high-level feature extraction in the integrated model. After data preprocessing, spatial and temporal features are extracted in CNN and recurrent autoencoder for the classification in fully connected networks. Empiric results show that the proposed model has better performances on multiple classification metrics and achieves preferable effect on anomaly detection.

This review presents the applications of biomass-derived porous graphitic carbon materials and their synthetic methods.

Nucleic acids hold promise as biomolecules for future applications in biomedicine and biotechnology. Their well-defined structures and compositions afford unique chemical properties and biological functions. Moreover, the specificity of hydrogen-bonded Watson-Crick interactions allows the construction of nucleic acid sequences with multiple functions. In particular, the development of nucleic acid probes as essential molecular engineering tools will make a significant contribution to advancements in biosensing, bioimaging and therapy. The molecular beacon (MB), first conceptualized by Tyagi and Kramer in 1996, is an excellent example of a double-stranded nucleic acid (dsDNA) probe. Although inactive in the absence of a target, dsDNA probes can report the presence of a specific target through hybridization or a specific recognition-triggered change in conformation. MB probes are typically fluorescently labeled oligonucleotides that range from 25 to 35 nucleotides (nt) in length, and their structure can be divided into three components: stem, loop and reporter. The intrinsic merit of MBs depends on predictable design, reproducibility of synthesis, simplicity of modification, and built-in signal transduction. Using resonance energy transfer (RET) for signal transduction, MBs are further endowed with increased sensitivity, rapid response and universality, making them ideal for chemical sensing, environmental monitoring and biological imaging, in contrast to other nucleic acid probes. Furthermore, integrating MBs with targeting ligands or molecular drugs can substantially support their in vivo applications in theranositics. In this review, we survey advances in bioanalytical and biomedical applications of rationally designed MBs, as they have evolved through the collaborative efforts of many researchers. We first discuss improvements to the three components of MBs: stem, loop and reporter. The current applications of MBs in biosensing, bioimaging and therapy will then be described. In particular, we emphasize recent progress in constructing MB-based biosensors in homogeneous solution or on solid surfaces. We expect that such rationally designed and functionalized MBs will open up new and exciting avenues for biological and medical research and applications.

Accurate and rapid load forecasting for industrial customers has been playing a crucial role in modern power systems. Due to the variability of industrial customers' activities, individual industrial loads are usually too volatile to forecast accurately. In this paper, a short-term load forecasting model for industrial customers based on the Temporal Convolutional Network (TCN) and Light Gradient Boosting Machine (LightGBM) is proposed. Firstly, a fixed-length sliding time window method is adopted to reconstruct the electrical features. Next, the TCN is utilized to extract the hidden information and long-term temporal relationships in the input features including electrical features, a meteorological feature and date features. Further, a state-of-the-art LightGBM capable of forecasting industrial customers' loads is adopted. The effectiveness of the proposed model is demonstrated by using datasets from different industries in China, Australia and Ireland. Multiple experiments and comparisons with existing models show that the proposed model provides accurate load forecasting results.

Using nanomaterials to mimic the function of protein enzymes is an interesting idea. Many nanomaterials have a similar size as enzymes and they also possess catalytic activity. Over the past decade, a surge of nanozyme work has emerged, likely due to the advancement in the synthesis and characterization of inorganic nanoparticles. Many typical enzymatic reactions mimicking oxidases, peroxidases, laccases, superoxide dismutases, and catalases have been realized by simple metal oxide and metal nanoparticles. In addition, small inorganic catalysts have been loaded in nanoparticles to create another type of nanozyme. The applications of nanozymes in biosensor design, environmental remediation, and therapeutics have been demonstrated. In this Topical Review, we briefly summarize the current status of the field and then focus our attention on some important problems faced by the field. These topics include developing better nanozymes with higher activity, better substrate selectivity, and engineering enzyme-like active sites. For practical applications, reliable methods for bioconjugation of nanozymes with affinity ligands need to be achieved, but not at the cost of losing the activity of nanozymes. Finally, fundamental mechanistic studies are needed to rationally design nanozymes and to obtain key insights into a few model systems.

Abstract Metal‐organic frameworks (MOFs) are of quite a significance in the field of inorganic‐organic hybrid crystals. Especially, MOFs have attracted increasing attention in recent years due to their large specific surface area, desirable electrical conductivity, controllable porosity, tunable geometric structure, and excellent thermal/chemical stability. Some recent studies have shown that carbon materials prepared by MOFs as precursors can retain the privileged structure of MOFs, such as large specific surface area and porous structure and, in contrast, realize in situ doping with heteroatoms (eg, N, S, P, and B). Moreover, by selecting appropriate MOF precursors, the composition and morphology of the carbon products can be easily adjusted. These remarkable structural advantages enable the great potential of MOF‐derived carbon as high‐performance energy materials, which to date have been applied in the fields of energy storage and conversion systems. In this review, we summarize the latest advances in MOF‐derived carbon materials for energy storage applications. We first introduce the compositions, structures, and synthesis methods of MOF‐derived carbon materials, and then discuss their applications and potentials in energy storage systems, including rechargeable lithium/sodium‐ion batteries, lithium‐sulfur batteries, supercapacitors, and so forth, in detail. Finally, we put forward our own perspectives on the future development of MOF‐derived carbon materials.