Benzyl alcohol oxidation and hydrogen generation over MoS2/ZnIn2S4 composite photocatalyst

Research on Chemical Intermediates - Owing to the simultaneous utilization of photogenerated electrons and holes, coupling light-driven hydrogen (H2) evolution reaction (HER) and biomass-derived...     

电缆槽对电缆间串扰的影响研究

电缆在使用时一般会铺设在电缆槽中,而电缆槽对铺设其中电缆之间的串扰会形成影响.针对电缆槽中电缆之间的串扰问题,提出一种结合多端口网络理论及分块级联思想的建模方法,使用基于波导格林函数的矩量法来提取电缆槽中电缆间的分布参数,并用虚拟节点理论将提取的分布参数矩阵转化为模型中的传输函数.通过与商业电磁仿真软件的仿真结果进行对比,验证了本文的模型及算法在分析电缆槽中电缆间的串扰问题时具有较好的精确度,最后运用此方法分析了电缆处于电缆槽中不同位置时的线间串扰大小,并结合铁路现场的实际情况对电缆槽中电缆的布线方式提出了建议.     

Multifunctional fibrous wound dressings for refractory wound healing

Refractory wounds have always been an important issue to healthcare systems, whose healing process is always delayed by multiple factors, including bacterial infections, chronic inflammation, and excessive exudates, etc. Employing multifunctional wound dressings is recognized as an effective strategy to deal with refractory wounds, which has yielded promising outcomes in recent years. Among these advanced wound dressings, fibrous dressings have gained growing attention due to their unique merits. Such wound dressings have demonstrated great potential in delivering theranostic agents, such as antibacterial agents, anti-inflammatory drugs, growth factors, and diagnostic probes, etc., for the purposes of accelerating wound healing. This paper reviews the development of multifunctional fibrous dressings and their applications in treating refractory wounds. The construction approaches of novel fibrous dressing with capabilities of antibacterial, anti-inflammation, exudate management and diagnosis were also introduced. Furthermore, the existing problems and challenges are also discussed briefly.     

Improved production of levoglucosan and levoglucosenone from acid-impregnated cellulose via fast pyrolysis

Cellulose - In this research, the production of levoglucosan (LG) and levoglucosenone (LGO) was improved from acid-impregnated cellulose via fast pyrolysis. Thermogravimetric and kinetic analysis...     

High-performance nonfused ring electron acceptor with a steric hindrance induced planar molecular backbone

Three nonfused ring electron acceptors (NFREAs) TTC6,TT-C8T and TT-TC8 were purposefully designed and synthesized.The molecular geometry can be adjusted by the steric hindrance of lateral substituents.According to the DFT calculations,from TTC6 to TT-C8T and TT-TC8,planarity of the molecular backbone is gradually improved,accompanying with the enhancing of intramolecular charge transfer effect.As for TT-TC8,the two phenyl substituents are almost perpendicular to the molecular backbone,which endues the acceptor with good solubility and suppresses it to form over-aggregation.Multidirectional regular molecular orientation and closer molecular stacking are formed in TT-TC8 film.As a result,TT-TC8 based devices afford the highest PCE of 13.13%,which is much higher than that of TTC6 (4.41%) and TT-C8T (10.42%) and among the highest PCE values based on NFREAs.     

Graphene Saturable Absorber Mirror for Mode-Locked Orange Pulse Laser

Journal of Russian Laser Research - We demonstrate a passively-mode-locked Pr3+: LuLiF4 laser operated at the 604 nm orange wave-length, using monolayer graphene as a saturable absorber. The...     

Observers Design for a Class of Nonlinear Stochastic Discrete-Time Systems

International Journal of Theoretical Physics - This paper mainly studied the observer design of Lipschitz stochastic discrete system. For the first time, generalized Lipschitz conditions are...     

基于深度学习的Stack Overflow问题帖分类方法

针对基于正则表达式和传统机器学习的分类方法分别存在模式手工提取困难和性能瓶颈的问题, 提出一种基于深度学习的问题帖分类方法, 采用深度文本挖掘模型TextCNN和融合注意力机制的TextRNN构建分类模型. 实验结果表明, 基于深度学习的方法在多数问题目的类别上的分类性能优于已有基准方法, 且使用的Adam优化器优于SGD优化器, 使用Glove预训练的词向量优于使用随机生成的词向量. 该方法以提问目的对帖子进行分类, 可为分析Stack Overflow(SO)上的帖子讨论主题增加新维度.     

基于GA_XGBoost模型的大学生科研能力预测问题研究

科学评价大学生科研创新能力对我国科研水平的提高具有重要意义.采用机器学习模型来预测大学生科研能力可以起到良好的效果,提出一种GAXGBoost模型来实现对大学生的科研能力预测.此模型是以Xgboost算法为基础,然后充分利用遗传算法的全局搜索能力自动搜索Xgboost最优超参数,避免了人为经验调参不准确的缺陷,最后采用精英选择策略以此确保每一轮都是最佳的进化结果.通过分析表明,所采用的GAXGBoost模型在大学生科研能力预测的结果中具有很高的精度,将此模型与Logistic Regression、Random Forest、SVM等模型进行对比,GAXGBoost模型的预测精度最高.