物质中芳环堆积的正确表达方式

通过一些晶体结构的实例指出，目前广泛采用的质心距小于0.4 nm的判据，在许多情况下并不能正确反映物质中的芳环堆积现象.为了正确判断在平行的芳香化合物之间是否存在近距离的芳环堆积现象，通过理论推算，给出了合理的判断依据：环平面的间距显著小于0.36 nm的范德华厚度，相邻芳环在平面法线方向的投影相互重叠或部分重叠，以上两者需同时满足.晶体学采用的是以晶胞为基础的斜坐标系，为了正确计算斜坐标系中平行平面的间距，利用转换矩阵，将斜坐标转换成归一化的笛卡儿坐标，在此基础上得出平行芳环之间面间距的正确计算方法.     

Self-energy and interaction energy of stacking fault in fcc metals calculated by embedded-atom method

The stacking fault energies of five fcc metals (Cu, Ag, Au, Ni and Al) with various quan-tivalences have been calculated by embedded-atom method (EAM). It indicated that the stacking fault energy is mainly determined by the metallic bond-energy and the lattice constant. Thus, monovalent fcc metals should have different stacking fault energies, contrary to Attree's conclusion. The interaction energy between stacking faults one I 111 I layer apart in a fcc metal is found to be 1/40-1/250 of its self-energy, while it becomes zero when the two stacking faults are two layers apart. The twin energy is just half of the energy of intrinsic stacking fault energy without the consideration of lattice relaxation and the energy of a single intrinsic stacking fault is almost the same as that of extrinsic stacking fault, which are consistent with the results from the calculation of Lennard-Jones force between atoms, but differ from Attree's result.     

重载智能堆垛装备刚度建模与结构优化

针对重载、高精度、高可靠性、长距离等货物搬运要求,提出重载智能堆垛装备的结构方案.通过对堆垛装备的工作原理及承载受力分析,提出基于V型滚轮接触力计算法的堆垛装备综合刚度求解方法,给出计算实例. 利用有限元方法求解V型导轨的接触刚度及堆垛装备的综合变形量,与理论计算结果进行对比,验证了理论方法的准确性. 通过建立基于刚度最大为优化目标的优化模型,对龙门立柱与三级货叉进行结构优化,得到立柱优化结构以及货叉最优的截面参数. 对堆垛装备进行实机运行及静态承载下的综合挠度检测. 结果表明,优化后的堆垛装备能够满足工程应用的要求.     

Determination of stacking fault energies in a high-Nb TiAl alloy at 298 K and 1273 K

The stacking fault energies of Ti-46A1-8.5Nb-0.2W alloy at 298 K and 1273 K were determined. The principle for the determination of the stacking fault energies is based on the fact that the stacking fault energy and the elastic interaction energy acting on the dissociated partial dislocations are equal. After the compress deformations with the strain of 0.2% at 298 K and 1273 K, and water quench to maintain the dislocation structures deformed at 1273 K, the dissociation distances between two partial dislocations were determined by weak beam transmission electron microscopy (WBTEM) technique. Based on these dissociation distances and the corresponding calculation method, the stacking fault energies were determined to be 77-81 mJ/m^2 at 298 K and to be 57-60mJ/m^2 at 1273 K respectively.     

卷积神经网络混合模型的心律失常分类算法

心律失常表现为不规则的心跳,心律失常类型的判断是心血管疾病早期预防和诊断的关键.为提高心律失常分类的准确率和速度,实现心律失常类型的自动识别,研究并提出了一种以卷积神经网络(convolutional neural network,CNN)为核心的7层混合模型结构.为保持心拍的完整性,根据R-R间期对心电信号进行动态分...     

The influence of temperature on stacking fault energy in Fe-based alloys

Temperature has great influence on the stacking fault energy (SFE). Both SFE and dr0/ d T for Fe-based alloys containing substitutional or interstitial atoms increase with increasing tempera-ture. Based on the thermodynamic model of SFE, the equation dr0/dT=drcn/dT+drseg/dT+drMG/dT and thoseexpressions for three items involved are established. The calculated dr0/dT is generally consistent with the experimental. The influence of chemical free energy on the temperature dependence of SFE is almost constant, and is obviously stronger than that of magnetic and segregation contributions. The magnetic transition and the segregation of alloying elements at stacking faults cause a decrease in SFE of the alloys when temperature increases; that is, drMG/dT<0 and dyseg/dT<0. Meanwhile, such an influence decreases with increasing temperature, except for the drseg/d 7 of Fe-Mn-Si alloys. With these results, the experimenal phenomena that the SFE of Fe-based alloys is not zero at the thermo-dynamically equilibrated tem     

基于半监督学习的脑电信号特征提取及识别

针对有监督学习容易造成未标记样本的浪费和手动特征提取容易导致信息丢失的问题,提出一种基于深层堆叠网络（DSN）的半监督特征学习方法,无监督特征学习的过程由多个受限玻尔兹曼机（RBM）的并行训练完成,将训练得到的参数用于DSN的输入权值初始化,再采用批量模式的梯度下降法进行监督微调。将所提方法用于运动想象脑电信号特征提取及识别,实验结果表明本文方法能够充分利用未标记样本中的隐含信息,有效提取脑电信号特征,识别结果优于共同空间模式（CSP）和深度信念网络（DBN）等算法,该方法可用于提高BCI系统中脑电信号的识别准确率。     

晶体结构中罕见的O-H…π堆积作用

Hinokiol是一种在植物中含量较少的二萜类化合物，对一些细菌有非常好的杀灭作用。Hinokiol从糙苏（Phlomis umbrosa Turcz）中首次分离出来，利用红外图谱和核磁图谱确定了其结构，并利用单晶X-衍射技术对其晶体结构和构象进行了分析，发现Hinokiol晶体结构中除存在氢键外还存在一种罕见的O-H…π堆积作用，这里对O-H…π堆积作用的特征进行了简单地描述。     

Li-Hopfield神经网络用于汉字字符识别

提出了一种Li-Hopfield网络,解决了Hopfield神经网络能量函数存在积分项、振荡收敛、计算复杂等问题,并继承了Hopfield网络的结构与Li网络的优点.同时发现基于部首拆分的汉字字符识别可以忽视甚至利用该网络的虚假稳定点,从而使样本存储量降为传统Hopfileld网络的1.44%,表明该算法有效且快速.     

端点自由DP应用于语音识别中的时间轴校正

本文介绍了一种新的时间轴校正方法,克服了传统端点固定DP在语音起止区间存在的固有误差,有效的提高了语音识别率。     

基于心搏聚类的Holter运动伪差段快速识别算法

为了能够在Holter分析系统中快速地标识出夹杂着运动伪差的时间段,提出一种基于心搏聚类的运动伪差段识别算法.先对心搏的QRS复合波面积进行直方图统计,得到心搏特征的主导分布;结合心搏形态,采用基于“轮廓限围”和“相关差值”的模板匹配算法完成心搏聚类;根据聚类结果得到的心搏凌乱特征曲线,自动识别出存在运动干扰的时间段.从医院采集了运动伪差比较明显的10例24 h Holter临床数据进行算法测试,实验表明：24 h的心电数据平均耗时不到1 s,识别结果与人工标注基本吻合.     

利用模式识别对柴油机进行故障诊断的一种方法

通过对柴油机的几种常见油管压力故障信号的分析 ,提出了一种提取模板向量和信息向量的方法 ,并且建立了五种模板库 ;利用欧式距离作为信息向量和模板向量的相似性测度 ,设计了一种基于最近邻准则的试探聚类算法来进行故障诊断 ;最后提出了一种可以通过人工干预、完善故障模板类以避免错误诊断的方法。实验结果证明这种方法可以有效的利用油管压力信号对柴油机进行故障诊断 ,当模板库逐步完善到一定程度时可以达到比较高的正确识别率     

Method to generate training samples for neural network used in target recognition

Training neural network to recognize targets needs a lot of samples.People usually get these samples in a non-systematic way,which can miss or overemphasize some target information.To improve this situation,a new method based on virtual model and invariant moments was proposed to generate training samples.The method was composed of the following steps:use computer and simulation software to build target object’s virtual model and then simulate the environment,light condition,camera parameter,etc.;rotate the model by spin and nutation of inclination to get the image sequence by virtual camera;preprocess each image and transfer them into binary image;calculate the invariant moments for each image and get a vectors’ sequence.The vectors’ sequence which was proved to be complete became the training samples together with the target outputs.The simulated results showed that the proposed method could be used to recognize the real targets and improve the accuracy of target recognition effectively when the sampling interval was short enough and the circumstance simulation was close enough.     

多层光网络实时选路算法

针对动态多层光网络(IP over WDM)的实时选路问题,提出了一种基于蚁群优化的区分综合选路算法．首先利用分层图模型,将传统的动态RWA问题转换为路由问题．然后,对该路由问题进行选路,通过将寻路蚂蚁放置在控制平面中,可以使控制平面的选路和数据平面的业务传输同步进行,从而为每一个动态到达的业务实时地进行选路; 并能综合考虑到路径的跳数和路径上链路的拥塞情况,有效降低业务阻塞率．最后,利用多类不同的蚂蚁对应不同粒度请求的业务,为不同的业务进行有区分的选路,使得高粒度请求的业务能绕开关键链路,从而在关键链路上承载更多的业务,进一步降低阻塞率．仿真实验表明,与其他算法相比,该算法具有更低的业务阻塞率和更好的业务阻塞公平性．     

A new acoustic model used in Chinese order recognition system for space teleoperating manipulator

０　ＩＮＴＲＯＤＵＣＴＩＯＮＩｎｏｒｄｅｒｔｏｄｅｖｅｌｏｐａＣｈｉｎｅｓｅｏｒｄｅｒｒｅｃｏｇｎｉｔｉｏｎｓｙｓｔｅｍｔｏｃｏｎｔｒｏｌｓｐａｃｅｔｅｌｅｏｐｅｒａｔｉｎｇｍａｎｉｐｕｌａｔｏｒｓｂｙｓｏｕｎｄ,ｉｔｉｓｖｅｒｙｉｍｐｏｒｔａｎｔ...     

数字技术和软件仿真相结合的心电图信号发生器

介绍了采用ATMEL89C2 0 5 1单片机作为微处理机 (MPU) ,将数字技术和软件仿真结合 ,研制的心电图信号发生器。它可以产生多种标准心电图信号波形 ,对仪器进行实时检验标定 ,也可作为模拟训练仪器     

数字技术和软件仿真相结合的心电图信号发生器

介绍了采用ATMEL89C2051单片机作为微处理机（MPU），将数字技术和软件仿真结合。研制的心电图信号发生器，它可以产生多种标准心电图信号波形，对仪器进行实时检验标定，也可作为模拟训练仪器。     

多深度层场景光场渲染的无重影条件

为避免光场渲染的结果因采样率不足而发生重影,提出了多深度层复杂场景的无重影光场渲染条件,证明了可由场景几何及摄像机角分辨率确定,不发生重影的最大摄像机间隔及最优深度,并在此基础上推导了两者间的关系。计算了不同摄像机间隔下的合成视图与标准视图间的峰值信噪比(PSNR),并以此衡量渲染质量。实验数据及人眼对重影发生与否的主观判定结论与本文的理论分析基本一致。     

新生心律失常的分析及病因探讨

本文分析了大学生新生查体中34例心律失常现象,其中以窦性心动过速和室心早搏较多见,以植物神经功能紊乱较多见(28例),82%).     

沉降预测的多层递阶时间序列模型研究

为了研究沉降模型中参数的时变特性，提出了多层递阶时间序列模型，该模型将沉降预测分为预测模型参数的预测和在此基础上的沉降预测两部分.采用参数递推公式计算沉降时间序列模型的参数序列，并根据计算出的参数序列构建其时间序列预测模型.依此类推，可建立多层参数的预测公式，直至认为参数不随时间变化为止，然后根据参数预测结果来预测后期沉降.实例分析和计算表明，多层递阶时间序列模型能较好地反映参数的时变特性，取得较精确的沉降预测结果.