线圈型全光纤偏振器的研究

本文采用等效电流法对线圈型全光纤偏振器进行了分析计算.对应于保偏光纤中两正交传输基模的消光比,提出了一种分析保偏光纤弯曲损耗的新方法,这种方法基于将保偏光纤两个不同偏振模式分别等效为两个普通圆光纤的基模,等效参数通过测量不同偏振模式的模场半径来确定.实验证明这种理论计算方法与实验结果一致.最终,我们研制出的线圈型光纤偏振器在1.31μm工作波长下,实测消光比达30dB以上,工作偏振模式插入损耗≤3dB.     

On sequential procedures for the point estimation of the mean of a normal population

The sequential procedures developed by Starr (1966, Ann. Math. Statist., 37, 1173–1185) for estimating the mean of a normal population are further analyzed. Asymptotic properties of the regret and first two moments of the stopping rules are studied and second-order approximations are derived.     

Case for projectile kinetic energy gain in stopping power studies

Under certain collision conditions, a swift ion projectile colliding with a target will gain rather than lose kinetic energy, contrary to the standard conception of stopping power. In this work, we consider the conditions for such a collision such that the energy loss is negative, that is, that there will be projectile kinetic energy gain. In particular, for a target initially in the ground state we find that the projectile gains kinetic energy only when charge exchange and de‐excitation processes are involved. This occurs when the electron affinity of the projectile is larger than the ionization potential of the target. Consequences of this effect are analyzed. © 2003 Wiley Periodicals, Inc. Int J Quantum Chem 94: 215–221, 2003     

纳米Zn-Co铁氧体的固相合成及电磁损耗特性研究

采用NH₄HCO₃与FeCl₃·6H₂O，Zn(NO₃)₂·6H₂O，Co(NO₃)₂·6H₂O进行室温固相反应制得前驱物，经微波加热处理后，进而热分解分别制得复合氧化物ZnFe₂O₄、CoFe₂O₄和Co_0.5Zn_0.5Fe₂O₄。由激光粒度分析仪、XRD和SEM表征：获得了颗粒分布比较均一、平均粒度为65 nm左右、立方晶系尖晶石结构的纳米铁氧体粉体。经测试样品的相对介电常数和相对磁导率后，研究了它们的电磁损耗特性。结果表明：Co_0.5Zn_0.5Fe₂O₄在100～1 800 MHz内比另外两种铁氧体具有更好的电磁损耗特性。     

Electron microscopic phase analysis of BN-thin films

A series of BN films was deposited by means of r.f. magnetron sputtering of a h-BN target onto Si(1OO) surfaces. Hereby, the substrate bias voltage was varied. Special interest is focussed to the influence of the deposition parameters on the orientation of the growing hexagonal BN film with respect to the substrate. For structural investigation, cross section samples were prepared. In addition to HRTEM and diffraction investigations, especially electron energy loss spectroscopy (EELS) was applied successfully for phase identification. For negative bias voltages of U _B =–300 V and U _B =–350V, we found a phase system consisting of a first-grown 25 nm thick layer of hexagonal structure with the c axis parallel to the substrate surface followed by the cubic phase.Dedicated to Professor Dr. rer. nat. Dr. h.c. Hubertus Nickel on the occasion of his 65th birthday     

龋病人牙釉质表面硬度及抗磨损性能测试

采用人工方法制成人牙釉质浅龋及使用抗龋措施的龋损釉质样本，分别测量各组釉质表面显微硬度，并以微动摩擦磨损试验机测试其抗磨性能，试验结果表明：各组实验组釉质较对照组表面硬度都有不同程度下降，氟化钠前处理会使釉质表达硬度显著下降，人工浅龋牙釉质的抗磨损性能不如正常釉质，使用两种抗龋措施对其抗磨性能没有明显改善。     

稀土变质热锻模具铸钢高温磨损性能的研究

研究了稀土（RE）变质热锻模具铸钢的高温磨损性能，并与热锻模具钢H13钢和3Cr2W8V钢进行对比，探讨了稀土元素的作用和热锻模具铸钢的高温磨损机理。结果表明：随着RE加入量的增加，热锻模具铸钢的磨损率先减后增，RE加入量在质量分数为0．05％时热锻模具铸钢具有最佳的高温磨损性能。RE变质热锻模其铸钢的高温耐磨性明最高于H13钢和3Cr2W8V钢。高温磨损机理为氧化磨损和氧化物的疲劳剥落，磨屑为块状的Fe2O3和Fe3O4。     

Neutral loss of amino acid residues from protonated peptides in collision-induced dissociation generates N- or C-terminal sequence ladders

The widespread occurrence of the neutral loss of one to six amino acid residues as neutral fragments from doubly protonated tryptic peptides is documented for 23 peptides with individual sequences. Neutral loss of amino acids from the N-terminus of doubly charged tryptic peptides results in doubly charged y-ions, forming a ladder-like series with the ions [M + 2H](2+) = y(max) (2+), y(max - 1) (2+), y(max - 2) (2+), etc. An internal residue such as histidine, proline, lysine or arginine appears to favor this type of fragmentation, although it was sometimes also observed for peptides without this structure. For doubly protonated non-tryptic peptides with one of these residues at or near the N-terminus, we observed neutral loss from the C-terminus, resulting in a doubly charged b-type ion ladder. The analyses were performed by Q-TOF tandem mass spectrometry, facilitating the recognition of neutral loss ladders by their 2+ charge state and the conversion of the observed mass differences into reliable sequence information. It is shown that the neutral loss of amino acid residues requires low collision offset values, a simple mechanistic explanation based on established fragmentation rules is proposed and the utility of this neutral loss fragmentation pathway as an additional source for dependable peptide sequence information is documented.     

Robust risk-taking under a sustainable constraint

We consider a model in which the representative investor makes optimal portfolio and consumption choices robust to ambiguity with a sustainable constraint. We find that the influences of ambiguity on risk-taking are two ways. Those are gambling when the risk-free interest rate is less than a critical value, but derisk when the risk-free interest rate is greater than the critical value. Moreover, the erosion of ambiguity on consumption is more substantial in constrained cases, and ambiguity will magnify welfare losses.     

增量式Huber-支持向量回归机算法研究

传统的面向支持向量回归的一次性建模算法中样本增加时,均需从头开始学习,而增量式算法可以充分利用上一阶段的学习成果。SVR的增量算法通常基于ε-不敏感损失函数,该损失函数对大的异常值比较敏感,而Huber损失函数对异常值敏感度低。所以在有噪声的情况下,Huber损失函数是比ε-不敏感损失函数更好的选择,在现实情况当中。基于此,本文提出了一种基于Huber损失函数的增量式Huber-SVR算法,该算法能够持续地将新样本信息集成到已经构建好的模型中,而不是重新建模。与增量式ε-SVR算法和增量式RBF算法相比,在对真实数据进行预测建模时,增量式Huber-SVR算法具有更高的预测精度。