基于CouchDB和Elastic Search的高性能化学结构搜索引擎与数据库的构建（英文）

计算机辅助的化学结构搜索在化学信息学中地位十分重要,本文设计了一套高性能的化学结构和化学数据搜索系统,称为DCAIKU. DCAIKU基于CouchDB无模式数据库和ElasticSearch基础架构构建,通过将结构相似性搜索变换为文字搜索实现了高性能和高灵活性的检索引擎:在满足化学信息存储的高灵活性条件下,仍然可以做到低延迟和高准确性,同时拥有良好的伸缩性,可以大规模并行化和集群化.     

机器学习辅助高性能化学信息数据库促进金属有机框架材料基气体吸附材料筛选

基于数据库化学结构搜索和机器学习快速筛选特定功能材料是近年的研究热点. 本文建立了基于MYSQL的高性能化学结构数据库,即MYDB. 数据库利用新的检索算法收集和存储了超过16万个金属有机框架材料,可以实现了高效检索和推荐. 测试结果显示MYDB能够在百万数量级的材料中实现快速高效的关键词搜索,并对相似结构提供实时推荐. 结合机器学习方法和材料数据库,训练了气体吸附模型,以确定一定热力学条件下金属有机框架材料对氩气和氢气的吸附能力. 结合MYDB数据库和机器学习算法训练出的模型能够支持大规模、低成本且方便快捷的结构筛选,从而推进计算材料研究领域中特定功能材料的发现.     

机器学习辅助高性能化学信息数据库促进金属有机框架材料基气体吸附材料筛选（英文）

基于数据库化学结构搜索和机器学习快速筛选特定功能材料是近年的研究热点.本文建立了基于MYSQL的高性能化学结构数据库,即MYDB.数据库利用新的检索算法收集和存储了超过16万个金属有机框架材料,可以实现了高效检索和推荐.测试结果显示MYDB能够在百万数量级的材料中实现快速高效的关键词搜索,并对相似结构提供实时推荐.结合机器学习方法和材料数据库,训练了气体吸附模型,以确定一定热力学条件下金属有机框架材料对氩气和氢气的吸附能力.结合MYDB数据库和机器学习算法训练出的模型能够支持大规模、低成本且方便快捷的结构筛选,从而推进计算材料研究领域中特定功能材料的发现.     

平面团簇稳定结构的蒙特卡罗树搜索

以平面团簇为例提出了一种结合结构识别和蒙特卡罗树技术搜索稳定结构的新方法.体系原子之间的相互作用由两类模型势能函数来描述:Lennard-Jones二体势函数与基于Lennard-Jones势的三体势函数.考虑可能的三角晶格碎片作为候选结构,引入编号策略对结构进行快速识别,并运用蒙特卡罗树搜索研究稳定结构随着原子数增大的演化过程;对于能量较低的候选结构,进一步采取局域优化来获得对应体系的稳定结构.计算表明,Lennard-Jones二体势函数对应的三角晶格团簇更稳定;在特定的参数下,三体势函数对应的六角晶格团簇更稳定.结合结构识别和蒙特卡罗树搜索可以对候选结构空间进行高效扫描,在较短时间内更容易搜索到稳定的团簇结构,并可以与第一原理计算结合实现材料的结构预测.     

绿色环保化学机械抛光液的研究进展

原子级加工制造是实现半导体晶圆原子尺度超光滑表面的有效途径.作为大尺寸高精密功能材料的原子级表面制造的重要加工手段之一,化学机械抛光(chemical mechanical polishing,CMP)凭借化学腐蚀和机械磨削的耦合协同作用,成为实现先进材料或器件超光滑无损伤表面平坦化加工的关键技术,在航空、航天、微电子等众多领域得到了广泛应用.然而,为了实现原子层级超滑表面的制备,CMP工艺中常采用的化学腐蚀和机械磨削方法需要使用具有强烈腐蚀性和高毒性的危险化学品,对生态系统产生了不可逆转的危害.因此,本文以绿色环保高性能抛光液作为对象,对加工原子量级表面所采用的化学添加剂进行分类总结,详尽分析在CMP过程中化学添加剂对材料表面性质调制的作用机理,为在原子级尺度下改善表面性质提供可参考的依据.最后,提出了CMP抛光液在原子级加工研究中面临的挑战,并对未来抛光液发展方向作出了展望,这对原子尺度表面精度的进一步提升具有深远的现实意义.     

一种新型通用化图像处理系统的设计

提出了一种基于FPGA+DSP的通用化嵌入式图像处理模块,可以用于不同的项目,只需要针对不同的项目设计不同的AD、DA和接口就可以了。该处理模块具有高性能、低功耗、低成本、可扩展、可重构、灵活性强等优点,可以大大降低开发成本和设计周期。     

基于毫米级单晶石墨烯的倍频器性能研究

石墨烯作为一种拥有高电子迁移率和高饱和速度的二维材料,在射频电子学领域具有很大的应用潜力,引起了人们广泛的研究兴趣.近些年随着化学气相沉积制备石墨烯技术的发展,高质量大尺寸的单晶石墨烯生长技术也愈加成熟.本文基于化学气相沉积生长的毫米级单晶石墨烯,在高介电常数介质上制备出高性能的石墨烯倍频器,并且对其倍频特性做了系统的研究.研究结果表明:在输入信号频率为1 GHz时,倍频增益可以达到-23.4 dB,频谱纯度可以达到94%.研究了不同漏极偏压以及输入信号功率下倍频增益的变化特性,随着漏极偏压以及输入信号功率的增加,倍频增益增加.对具有不同跨导和电子空穴电导对称性的器件的倍频增益和频谱纯度随输入信号频率f_(in)的变化关系进行了研究.结果表明,跨导对于倍频增益影响显著,在f_(in)=1 GHz时器件的频谱纯度差别不大,均大于90%,但是随着f_(in)增加至4 GHz,电子空穴电导对称性较差的器件频谱纯度下降至42%,电子空穴电导对称性较好的器件仍能保持85%的频谱纯度.这是电子空穴电导对称性和电子空穴响应速度共同作用的结果.本文的研究结果对于高性能石墨烯倍频器设计具有一定的指导意义.     

高性能Ag/AgI红外空芯光纤的研究

采用液相化学镀膜法研制了内面镀有银膜和碘化银膜(Ag/AgI)的高性能空芯红外光纤.分析了绿色导航光和红外激光同时低损耗传输的光纤结构参量.长度1 m、内直径0.7 mm的Ag/AgI空芯光纤不仅在红外有低损耗传输特性,而且首次在波长530 nm处获得了7 dB/m的低传输损耗特性.满足可见光作为导航光的同轴传输要求.通过分析和改进实验工艺参量,有效降低Ag和AgI薄膜的表面粗糙度.给出了高性能Ag/AgI空芯光纤的介质镀膜的具体工艺参量.     

大尺度分层介质电特性参数的反演方法研究

针对大尺度分层介质电磁场散射回波数据, 提出了一种用于分层介质特性参数反演的新方法, 实现了分层介质的介电常数、电导率和介质厚度的同时反演. 方法首先将这些参数的反演转化为最优化问题进行求解, 然后应用模拟退火算法搜索最优解, 充分利用算法的全局寻优能力, 同时改善了算法的搜索控制策略, 使算法在最优解的搜索过程中可以自适应地调整搜索步长, 提高了算法的搜索效率. 结果表明, 本文方法可以实现大尺度分层介质特性参数的准确反演, 且具有较强的抗噪声能力. 该方法可以应用于火星/月球 雷达探测回波数据的反演以及地下分层结构特性的分析. 关键词： 电磁散射 分层介质 模拟退火算法 最优化     

氢稀释对纳米晶硅薄膜晶化特性的影响及薄膜生长机理

以SiH4与H2作为前驱气体,采用射频等离子增强化学气相沉积技术制备了纳米晶硅薄膜.利用Raman散射和红外吸收光谱等技术,对不同氢稀释比条件下薄膜的微观结构和键合特性进行了研究.结果表明,随着氢稀释比增加,薄膜的晶化率明显提高,而氢稀释比过高时,薄膜晶化率呈现减少趋势.红外吸收光谱分析表明,纳米晶硅薄膜中氢的键合模式与薄膜的晶化特性密切相关.随着氢稀释比增加,薄膜中整体氢含量和SiH2键合密度明显减少,而在高氢稀释比条件下,氢稀释比增加导致薄膜中SiH2键合密度和整体氢含量增加.     

Dynamics of a macroscopic model characterizing mutualism of search engines and web sites

We present a model to describe the mutualism relationship between search engines and web sites. In the model, search engines and web sites benefit from each other while the search engines are derived products of the web sites and cannot survive independently. Our goal is to show strategies for the search engines to survive in the internet market. From mathematical analysis of the model, we show that mutualism does not always result in survival. We show various conditions under which the search engines would tend to extinction, persist or grow explosively. Then by the conditions, we deduce a series of strategies for the search engines to survive in the internet market. We present conditions under which the initial number of consumers of the search engines has little contribution to their persistence, which is in agreement with the results in previous works. Furthermore, we show novel conditions under which the initial value plays an important role in the persistence of the search engines and deduce new strategies. We also give suggestions for the web sites to cooperate with the search engines in order to form a win–win situation.     

Mining combinatorial data in protein sequences and structures

Combinatorial searches of structural and physical chemical properties involving the components of libraries of dipeptide, tripeptide and tetra peptide fragments were carried out in the Protein Data Bank and the SwissProt databases. The properties investigated are structural propensities, co localization of peptide fragments in protein sequences, interactions between peptide fragments in close structural proximity and the participation of physical chemical profiles in the distribution of structural motifs among peptide fragments. The results obtained for each combinatorial search in the study are classified according to the structural motifs alpha-helix, beta-sheet, reverse turn I and reverse turn II. The application of combinatorial data mined in protein databases to the design of new peptide libraries is discussed. The present findings have implications for the study of protein structure which are also discussed.     

Efficiency bounds for nonequilibrium heat engines

We analyze the efficiency of thermal engines (either quantum or classical) working with a single heat reservoir like an atmosphere. The engine first gets an energy intake, which can be done in an arbitrary nonequilibrium way e.g. combustion of fuel. Then the engine performs the work and returns to the initial state. We distinguish two general classes of engines where the working body first equilibrates within itself and then performs the work (ergodic engine) or when it performs the work before equilibrating (non-ergodic engine). We show that in both cases the second law of thermodynamics limits their efficiency. For ergodic engines we find a rigorous upper bound for the efficiency, which is strictly smaller than the equivalent Carnot efficiency. I.e. the Carnot efficiency can be never achieved in single reservoir heat engines. For non-ergodic engines the efficiency can be higher and can exceed the equilibrium Carnot bound. By extending the fundamental thermodynamic relation to nonequilibrium processes, we find a rigorous thermodynamic bound for the efficiency of both ergodic and non-ergodic engines and show that it is given by the relative entropy of the nonequilibrium and initial equilibrium distributions. These results suggest a new general strategy for designing more efficient engines. We illustrate our ideas by using simple examples.     

The role of CALYPSO in the discovery of high-T_c hydrogen-rich superconductors

Hydrogen-rich compounds are promising candidates for high-Tc or even room-temperature superconductors. The search for high-Tc hydrides poses a major experimental challenge because there are many known hydrides and even more unknown hydrides with unusual stoichiometries under high pressure. The combination of crystal structure prediction and first-principles calculations has played an important role in the search for high-Tc hydrides, especially in guiding experimental synthesis. Crystal structure AnaLYsis by Particle Swarm Optimization(CALYPSO) is one of the most efficient methods for predicting stable or metastable structures from the chemical composition alone. This review summarizes the superconducting hydrides predicted using CALYPSO. We focus on two breakthroughs toward room-temperature superconductors initiated by CALYPSO: the prediction of high-Tc superconductivity in compressed hydrogen sulfide and lanthanum hydrides, both of which have been confirmed experimentally and have set new record Tc values. We also address the challenges and outlook in this field.     

Towards a high-accuracy kinetic database informed by theoretical and experimental data: CH3 + HO2 as a case study

The reliability of predictive simulations for advanced combustion engines depends on the availability of accurate data and models for thermochemistry, chemical kinetics, and transport. In that regard, accurate data are critically important for both their direct use in predictive simulations and for benchmarking improved theoretical methodologies that can similarly produce accurate data for predictive simulations. The use of informatics-based strategies for the determination of accurate thermochemical data with well-defined uncertainties, e.g. the Active Thermochemical Tables (ATcT), has revolutionized the field of thermochemistry–providing thermochemical data of unprecedented accuracy for predictive combustion simulations and has served as a key enabler of ab initio electronic structure methodologies of equally impressive accuracy. Clearly, pursuit of informatics-based analogs in chemical kinetics would be similarly worthwhile. Here, we present results and analyses for the kinetics of CH₃ + HO₂ reactions that demonstrate some key elements of any approach to developing analogs for kinetics, including: the importance of raw data for quantifying the information content of experimental data, the utility of theoretical kinetics calculations for constraining experimental interpretations and providing an independent data source, and the subtleties of target data selection for avoiding unphysical parameter adjustments to match data affected by structural uncertainties. Notably, we find the optimization performed here using the MultiScale Informatics (MSI) approach applied to carefully selected (mostly raw) experimental data yields an opposite temperature dependence for the channel-specific CH₃ + HO₂ rate constants as compared to a previous rate-parameter optimization. While both optimization studies use the same theoretical calculations to constrain model parameters, only the present optimization, which incorporates theory directly into the model structure, yields results that are consistent with theoretical calculations.     

System of HPC content archiving

This work is aimed to develop a system, that will effectively solve the problem of storing and analyzing files containing text data, by using modern software development tools, techniques and approaches. The main challenge of storing a large number of text documents defined at the problem formulation stage, have to be resolved with such functionality as full text search and document clustering depends on their contents. Main system features could be described with notions of distributed multilevel architecture, flexibility and interchangeability of components, achieved through the standard functionality incapsulation in independent executable modules.     

A Structural Entropy Measurement Principle of Propositional Formulas in Conjunctive Normal Form

The satisfiability (SAT) problem is a core problem in computer science. Existing studies have shown that most industrial SAT instances can be effectively solved by modern SAT solvers while random SAT instances cannot. It is believed that the structural characteristics of different SAT formula classes are the reasons behind this difference. In this paper, we study the structural properties of propositional formulas in conjunctive normal form (CNF) by the principle of structural entropy of formulas. First, we used structural entropy to measure the complex structure of a formula and found that the difficulty solving the formula is related to the structural entropy of the formula. The smaller the compressing information of a formula, the more difficult it is to solve the formula. Secondly, we proposed a λ-approximation strategy to approximate the structural entropy of large formulas. The experimental results showed that the proposed strategy can effectively approximate the structural entropy of the original formula and that the approximation ratio is more than 92%. Finally, we analyzed the structural properties of a formula in the solution process and found that a local search solver tends to select variables in different communities to perform the next round of searches during a search and that the structural entropy of a variable affects the probability of the variable being flipped. By using these conclusions, we also proposed an initial candidate solution generation strategy for a local search for SAT, and the experimental results showed that this strategy effectively improves the performance of the solvers CCAsat and Sparrow2011 when incorporated into these two solvers.     

A numerical study of the autoignition of dimethyl ether with temperature inhomogeneities

The autoignition of dimethyl ether (DME) with temperature inhomogeneities is investigated by one-dimensional numerical simulations with detailed chemistry at high pressure and a constant volume. The primary purpose of the study is to provide an understanding of the autoignition of DME in a simplified configuration that is relevant to homogeneous charge compression ignition (HCCI) engines. The ignition structure and the negative temperature coefficient (NTC) behaviour are characterised in a homogeneous domain and one-dimensional domains with thermal stratification, at different initial mean temperatures and length scales. The thermal stratification is shown to strongly affect the spatial structure and temporal progress of ignition. The importance of diffusion and conduction on the ignition progress is assessed. It is shown that the effects of molecular diffusion decay relative to those of chemical reaction as the length-scale increases. This is to be expected, however the present study shows that these characteristics also depend on the mean temperature due to NTC behaviour. For the range of conditions studied here, which encompass a range of stratification length scales expected in HCCI engines, the effects of molecular transport are found to be small compared with chemical reaction effects for mean temperatures within the NTC regime. This is in contrast to previous work with fuels with single-stage ignition behaviour where practically realisable temperature gradients can lead to molecular transport effects becoming important. In addition, thermal stratification is demonstrated to result in significant reductions of the pressure-rise rate (PRR), even for the present fuel with two-stage ignition and NTC behaviour. The reduction of PRR is however strongly dependent on the mean initial temperature. The stratification length-scale is also shown to have an important influence on the pressure oscillations, with large-amplitude oscillations possible for larger length scales typical of integral scales in HCCI engines.     

High-frequency modes in a two-dimensional rectangular room with windows

We examine a two-dimensional model problem of architectural acoustics on sound propagation in a rectangular room with windows. It is supposed that the walls are ideally flat and hard; the windows absorb all energy that falls upon them. We search for the modes of such a room having minimal attenuation indices, which have the expressed structure of billiard trajectories. The main attenuation mechanism for such modes is diffraction at the edges of the windows. We construct estimates for the attenuation indices of the given modes based on the solution to the Weinstein problem. We formulate diffraction problems similar to the statement of the Weinstein problem that describe the attenuation of billiard modes in complex situations.