一种求解约束优化问题的改进粒子群算法及其应用

针对复杂约束优化问题,提出一种改进的粒子群方法。该粒子群算法对于不满足约束条件的粒子实行全概率接收,但令其目标函数值同为一个很小的常数,以保持粒子的多样性并使最优解在可行域内。另外,在PSO算法的基础上,使惯性权值按对数规律单调递减,同时引进选择遗传算子,以增强其全局寻优性能。数值实验表明:与PSO算法和一些其它优化算法相比,改进算法具有较强的寻优能力和寻优效率。工程应用表明,改进算法具有一定的优越性。     

直喷式柴油机燃烧过程及碳烟生成的初步计算

依据已简化的包含多环芳烃生成的正庚烷气相详细化学反应机理,基于专业流体解析软件FLUENT6.3对直喷式柴油机压缩、膨胀冲程内的自燃和燃烧过程进行了动态模拟,得到了缸内平均温度、主要反应物以及中间产物多环芳烃生成量随转角变化的特性曲线.在数值上,计算得到以苯环质量为主的多环芳烃总生成量水平比试验测量结果约高出1~2个量级,但是在随转角变化的特性上,计算与试验测量曲线定性趋势一致.针对特定转角时刻主要火焰结构参数的截面分布,分析研究了柴油机缸内化学反应特性与流体流动相互作用下的空间分布特征.通过耦合详细碳烟粒子动力学模型,对碳烟表象尺寸进行了初步模拟.     

Field Dependence of π-Band Superconducting Gap in MgB2 Thin Films from Point-Contact Spectroscopy

We present the results of point-contact spectroscopy measurements on High-quality epitaxial MgB₂ thin films with injected current along the c-axis. The temperature and field dependences of π-band properties with the field parallel to (H_⊥) or perpendicular to (H_|| the c-axis are investigated in detail. When a magnetic field is applied, either parallel or perpendicular to the c-axis, the density of the quasiparticle state (DOS) of the π-band proliferates quickly with increasing field, while the gap amplitude of the π-band decreases slowly, which is different from the recent theoretical calculations, showing a field dependent competition between the interband scattering and the pair-breaking effects.     

Generation of Long-Distance Entanglement in Spin System

We propose two schemes to produce long-distance entanglement in a spin chain. The first is based on a controllable interaction system, one starts from an entangled kernel and adds weaken interaction spins to the boundary sites step by step, then the entanglement will be extended longer and longer and its value is equal to that of its kernel. The second is based on a uniform interaction （J） system with a bulk magnetic field （B） that is absent for the boundary qubits, as long as B/J 〉 5, one can obtain near perfect long distance entanglement. Ultra-low temperature is needed in both schemes.     

混流泵三元叶片优化设计方法研究

以叶片角的分布为优化变量,叶轮效率为优化目标,将三维粘性流动分析与神经网络相结合对混流泵叶片进行了优化设计。在设计空间内根据实验设计理论安排神经网络的训练样本,选择径向基函数网络建立设计变量和目标函数间复杂的响应关系,利用遗传算法进行优化。针对研究对象设计变量较多、变化范围较大的问题,提出子系统优化的初步思想,即:将优化变量进行分类,对每类变量建立子系统,分别在子系统内进行局部优化,然后再考虑子系统之间的相互影响进行全局优化。对一混流泵的三元扭曲叶片进行优化,叶轮效率明显提高,从而验证了该方法的简便性与有效性。     

二次微分和傅里叶解卷积技术的表观分辨率研究

二次微分和解卷积技术都可以提高红外光谱的表观分辨率 ,从而可以分辨谱图中重叠的峰。文章用二次微分和解卷积技术对 2 十二烷基TCNQ ,2 十五烷基TCNQ和 2 十八烷基TCNQ的红外光谱进行了处理 ,发现CH2 反对称和对称伸缩振动区域的两对吸收峰的强度与脂肪链的长度有关 ,并从实验上证实了解卷积技术比二次微分技术具有更强的表观分辨能力     

液晶面板取向层与液晶层界面的电荷累积效应分析

TN模式的电荷累积缺陷,是液晶显示器生产中一个相当棘手问题。由于累积电荷量难于定量测量来进行研究,所以该问题一直悬而未决。着重分析了液晶显示器电荷累积缺陷,发现当它们介电常数与电导率的比值相差越大,电荷积累值越大;当取向层的介电常数与电导率的比值和液晶的不相等时,二者的交界面将产生电荷积聚;并且发现当取向层的介电常数与电导率的比值和液晶的相等时,二者的交界面将不产生电荷积聚,图像残留、显示速度减慢等问题也就可以消除。     

实际气体多变功计算模型研究

本文研究了两种实际气体多变功的直接积分数值求解模型,并编制了相应的计算程序.通过与平均过程指数法求得的多变功结果对比发现,直接积分求解模型较平均过程指数法计算得到的多变功具有更高的精度.当压比较高时,平均指数法得到的多变功误差明显增大甚至超过了4%,而直接积分模型计算得到的多变功在高压比下仍具有较高精度.     

Amplified Spontaneous Emission of Organic Pyridinium Dye doped Polymeric Waveguide

An organic dye salt trans-4-[p-（N-hydroxyethyl-N-methylamino）styryl]-N-methylpyridinium iodide （ASPI） is doped with an electron transport organic molecule tris（8-hydroxyquinoline） aluminium （Alq3） in a host matrix of poly（methylmethacrylate） （PMMA）, and the amplified spontaneous emission （ASE） is studied. By efficient Forster energy transfer from Alq3 to ASPI, it is demonstrated that the ASE threshold of ASPI：Alq3：PMMA waveguide （about 11 μJ/pulse） is much lower than that of ASPI：PMMA system （about 38 μJ/pulse）. Meanwhile, the peak position of ASE can be controlled by the effect of film thickness on waveguide modes. We show that the ASE peak position can be tuned over 37nm. These characteristics indicate the ASPI：Alq3 system as a promising gain medium for optical amplifiers and organic semiconductor lasers.     

重离子辐照带有ECC的65 nm SRAM器件“伪多位翻转”特性研究

为了提高纠错编码（ECC）的有效性,先进的静态随机存储器（SRAM）多采用位交错结构。但是,在没有物理版图信息的情况下,位交错设计使得从辐照测试数据中提取出多单元翻转（MCU）变得更加困难。运用Bi离子辐照带有ECC的65 nm SRAM器件,研究了该款器件在重离子辐照下的敏感性。为"伪多位翻转（FMBU）"以及MCU的数据分析提供了理论指导和帮助,完善了判别MCU的基本法则。除此之外,研究结果表明,ECC的汉明编码对于纳米器件的效果不够理想。在未来的空间应用中,需考虑更高层次的编码算法来抵抗单粒子翻转。In order to improve the robustness of error-correcting codes (ECC), modern static random access memory (SRAM) always use bit-interleaving structure. However, in the absence of physical layout information, the bit-interleaving design makes it more difficult to extract the multiple-cell upset (MCU) from the test data. In this paper, the sensitivity of Bi ion irradiation was investigated in a 65 nm technology SRAM with ECC. The experimental results provide a theoretical guidance and help for the fake multiple-bit upset (FMBU) and MCU data analyzing, which improve and perfect the basic rules extracting MCU from the test data. In addition, the results show that the performance of hamming encoding is not ideal in Nano scale SRAM. In the future of space applications, it is necessary to consider more advanced algorithms to against SEU.