融合图卷积神经网络和注意力机制的PM2.5小时浓度多步预测

PM_2.5小时浓度多为单步预测。为实现PM_2.5小时浓度的多步预测，基于“编码器-解码器”的序列-序列预测（Seq2Seq）模型，集合图卷积神经网络提取非欧式空间数据特征的能力以及注意力机制自适应关注特征的能力，提出了融合图卷积神经网络和注意力机制的PM_2.5小时浓度多步预测（GCN_Attention_Seq2Seq）模型。并与Seq2Seq模型和使用了图卷积神经网络、未使用注意力机制的GCN_Seq2Seq模型进行了对照，以2015—2016年北京市22个空气质量监测站点的空气质量数据为样本进行实例验证，结果表明，Seq2Seq模型和图卷积神经网络（GCN）可对PM_2.5小时浓度数据的时空依赖进行有效建模，注意力机制有助于减缓多步预测中的预测精度衰减，提升PM_2.5小时浓度多步预测的精度。GCN_Attention_Seq2Seq模型可有效应用于多种长度的PM_2.5浓度预测窗口。     

基于RBM的神经网络时间序列预测

为了克服神经网络依赖初始化结果,泛化能力不强的缺点,提出一种基于受限玻尔兹曼机(RBM)的神经网络模型.利用无监督学习方法优化神经网络的初始权值和阈值,将RBM与神经网络融合起来,模型与时间序列神经网络做实验对比,结果表明,基于受限的玻尔兹曼机的神经网络模型优于神经网络预测模型,模型可以提高预测的精准度,具有一定的应用意义.     

考虑防御系统延迟使其预测误差最大的突防弹道设计

对中段多脉冲机动突防弹道的设计问题进行了研究。采用的原理与方法有别于成熟的航天器多脉冲变轨,而是基于路径规划的思想对弹道设计问题简化。考虑敌方防御系统延迟,改进多脉冲点火模型,并基于变射面的思想对关机点参数进行了设计及优化。然后,从使敌方预测误差最大的角度提出评价函数,并利用遗传算法进行优化。最后,给出了一种同时满足导弹机动突防与打击精度要求的多脉冲弹道设计方法。从模型的可行性,方法的有效性、灵活性、迭代效率及精度等方面进行了仿真验证分析。在PC机仿真中,20 s内就设计出一条保证打击精度的中段突防弹道。结果表明,建立的模型是可行的,提出的突防弹道设计方法是快速有效的。     

An integrative in silico approach to the structure of Omp33-36 in Acinetobacter baumannii

Omp33-36 in A. baumannii, a bacterium causing serious nosocomial infections, is a virulence factor associated with the pathogen metabolic fitness as well as its adherence and invasion to human epithelial cells. This protein is also involved in interaction of the bacteria with host cells by binding to fibronectin. Moreover, Omp33-36 renders cytotoxicity to A. baumannii in addition to inducing apoptosis and modulation of autophagy. In the present study, an integrated strategy is launched to pierce into the 3D structure of Omp33-36 protein. The signal peptide within the sequence was determined, then, topology as well as secondary and tertiary structures of the protein were predicted. The mature protein assigned as a 14-stranded barrel in which residues 1–19 is removed as signal peptide. The obtained 3D models were evaluated in terms of quality; and then, served as queries to find similar protein structures. The hits were analyzed regarding topology among which 14-stranded were considered. The most qualified model was refined and then its sequence aligned to its counterpart similar structure protein (CymA from Klebsiella oxytoca). The determined structure of Omp33-36 could justify its porin function and carbapenem-resistance associated with the loss of this protein.     

Temperature dependence of the properties of ternary systems ethanol + water + an aliphatic alkane (C6-C9)

The measures and calculation of different properties such as refractive index, density, speed of sound, excess molar volume, and isentropic compressibility of the ternary heterogeneous compounds by ethanol + water + (n-hexane, n-heptane, n-octane, n-nonane) have been performed in the range 288.15–323.15 K and atmospheric pressure. Attending to the accurate results of these models, the equation of state enclosing mixing rules is indicated as a simple estimation of the procedures of these properties for this kind of multicomponent systems.     

A Dynamic Anomaly Detection Approach Based on Permutation Entropy for Predicting Aging-Related Failures

Software aging is a phenomenon referring to the performance degradation of a long-running software system. This phenomenon is an accumulative process during execution, which will gradually lead the system from a normal state to a failure-prone state. It is a crucial challenge for system reliability to predict the Aging-Related Failures (ARFs) accurately. In this paper, permutation entropy (PE) is modified to Multidimensional Multi-scale Permutation Entropy (MMPE) as a novel aging indicator to detect performance anomalies, since MMPE is sensitive to dynamic state changes. An experiment is set on the distributed database system Voldemort, and MMPE is calculated based on the collected performance metrics during execution. Finally, based on MMPE, a failure prediction model using the machine learning method to reveal the anomalies is presented, which can predict failures with high accuracy.     

基于不确定理论的铁路货运需求预测

合理预测铁路货运需求是铁路管理部门建设、运营等决策基础。为应对铁路货运需求的复杂变化,基于Pearson相关性分析方法筛选出铁路货运需求的七个具有关键影响的因素,并结合不确定理论建立不确定多元线性回归模型,相应的铁路货运预测结果由传统单一值变成可能的需求区间范围,更加符合处于不确定环境下的铁路货运需求实际情况。选取国家统计局2004~2016年相关数据进实证研究,并与回归模型以及BP模型的预测结果对比分析,实验表明不确定多元线性回归的预测结果更加精确。     

Structure Prediction of Binary Pernitride MN2 Compounds (M=Ca,Sr, Ba,La, and Ti)

Metal‐pernitride compounds belong to a class of chemical systems in which both the complex ions and the non‐bonding electrons may play roles in the formation of their modified crystalline structures. To investigate this issue, the energy landscapes of pernitrides of metals with different maximum valence (M=Ca, Sr, Ba, La, and Ti) were globally explored on the ab initio level at standard and high pressures, thereby yielding possible (meta)stable modifications in these systems together with information on how the landscape changed as function of the valence of the metal cation. For all of the systems in which no compounds had been synthesized so far, we predicted the existence of kinetically stable modifications that should, in principle, be experimentally accessible. In particular, TiN₂ should crystallize in a new structure type, TiN₂‐I.     

Concurrent combined verification: reducing false positives in automated NMR structure verification through the evaluation of multiple challenge control structures

Automated structure verification using ¹H NMR data or a combination of ¹H and heteronuclear single‐quantum correlation (HSQC) data is gaining more interest as a routine application for qualitative evaluation of large compound libraries produced by synthetic chemistry. The goal of this automated software method is to identify a manageable subset of compounds and data that require human review. In practice, the automated method will flag structure and data combinations that exhibit some inconsistency (i.e. strange chemical shifts, conflicts in multiplicity, or overestimated and underestimated integration values) and validate those that appear consistent. One drawback of this approach is that no automated system can guarantee that all passing structures are indeed correct structures. The major reason for this is that approaches using only ¹H or even ¹H and HSQC spectra often do not provide sufficient information to properly distinguish between similar structures. Therefore, current implementations of automated structure verification systems allow, in principle, false positive results. Presented in this work is a method that greatly reduces the probability of an automated validation system passing incorrect structures (i.e. false positives). This novel method was applied to automatically validate 127 non‐proprietary compounds from several commercial sources. Presented also is the impact of this approach on false positive and false negative results. Copyright © 2012 John Wiley & Sons, Ltd.