分段对称反向高斯-约当消元法及其应用

求解变系数方程的高斯消元法与高斯-约当消元法计算原理类似、问题相近,但前者计算速度高于后者.提出分段对称反向高斯-约当消元法,其中包括根据系数矩阵结构特点构成特殊增广阵,以展示和应用元素的变化规律,并分段对上下三角元素消元以大大提高计算效率.对矩阵下三角元素正向消元及对称计算可简化所有下三角元素计算,而对上三角元素反向消元可再省略所有上三角元素计算,而取倒后的对角元素作为规格化因子可大大减少除法计算.根据单位矩阵结构特点,对其规格化或对系数矩阵上下三角元素消元时均仅计算部分对角元素和下三角元素可进一步提高计算效率.所有元素均用四角规则计算而无需计算公式以简化计算和编程.新方法大大减少了高斯-约当消元法中元素的计算,且原理简单、易于编程,可快速求解各种变系数方程,还可利用元素对称性求解常系数的节点阻抗矩阵.与高斯消元法和高斯-约当消元法相比,新方法计算速度大大提高.     

High-voltage surge protection by a varistor-filled air gap

The electric discharge across a varistor granule filled air gap under a fast-rising voltage pulse was investigated for surge protection applications. The effects of temperature and pressure on the arc and the electrical conduction were analyzed by the characteristic changes in voltage waveforms triggered by a fast-rising high voltage pulse. In addition to the gap size, experimental results show that competing mechanisms among arc conduction, conduction through the varistor granule network, thermionic emission from Joule heating at granule-to-granule contact points, and the magnitude of the switching voltage dictate the maximum surge protection voltage for the filled air gap. Experimental evidence indicated that accumulated degradation was created at small contact points between varistor granules by repetitive assaults from longer duration, high voltage pulses. The uniqueness of using varistor over other dielectric granules in an air gap for surge protection is identified and discussed.     

Spatial distribution of silica fillers in phase-separated rubber blends investigated by three-dimensional elemental mapping

The distribution of nano-sized silica in binary rubber blends is characterized by scanning transmission electron microscopy (STEM) tomography combined with energy dispersive X-ray spectrometry (EDX). 3D distribution of silica is visualized by STEM-EDX tomography with the tilt-series of silicon elemental maps, while the phase-separated morphologies of polyisoprene rubber (IR) and styrene-butadiene rubber (SBR) are visualized by STEM-tomography in high-angle-annular-dark field (HAADF) mode. The combination of STEM-EDX and STEM-HAADF tomography enables us to determine the distribution of silica between the two rubber phases quantitatively even with high contents of silica up to 70 phr (weight parts per hundred rubber). It is found that silica is preferentially distributed in the SBR phase, but it is also distributed in the IR phase when the IR fraction in the total rubber components is higher than 40 wt%. The preferential distribution of silica in the SBR phase improves the dispersion of the IR domains. This is the first use of this technique for a multicomponent polymer system, showing the advantage to characterize the complicated multicomponent polymer composite morphologies.     

CAN总线伺服控制卡的研制应用

本文结合清华大学THBIP-I型仿人机器人样机研制，开发设计了基于CAN总线的仿人机器人关节伺服控制卡。文中详细描述了该控制卡的硬件结构原理和系统软件协议，以及伺服卡在样机上的技术实验，并提出了进一步完善发展的思路。     

Nonparametric Traversability Estimation in Partially Occluded and Deformable Terrain

Terrain traversability estimation is a fundamental requirement to ensure the safety of autonomous planetary rovers and their ability to conduct long‐term missions. This paper addresses two fundamental challenges for terrain traversability estimation techniques. First, representations of terrain data, which are typically built by the rover's onboard exteroceptive sensors, are often incomplete due to occlusions and sensor limitations. Second, during terrain traversal, the rover‐terrain interaction can cause terrain deformation, which may significantly alter the difficulty of traversal. We propose a novel approach built on Gaussian process (GP) regression to learn, and consequently to predict, the rover's attitude and chassis configuration on unstructured terrain using terrain geometry information only. First, given incomplete terrain data, we make an initial prediction under the assumption that the terrain is rigid, using a learnt kernel function. Then, we refine this initial estimate to account for the effects of potential terrain deformation, using a near‐to‐far learning approach based on multitask GP regression. We present an extensive experimental validation of the proposed approach on terrain that is mostly rocky and whose geometry changes as a result of loads from rover traversals. This demonstrates the ability of the proposed approach to accurately predict the rover's attitude and configuration in partially occluded and deformable terrain.     

Negative bias illumination stress assessment of indium gallium zinc oxide thin‐film transistors

Electrical performance stability of indium gallium zinc oxide (IGZO) thin‐film transistors (TFTs) is evaluated under negative bias illumination stress (NBIS). A bottom‐gate IGZO TFT whose top surface is passivated with zinc tin silicon oxide (ZTSO) exhibits a dramatic improvement in NBIS stability compared with that of an unpassivated, bottom‐gate IGZO TFT. Oxygen chemisorption/desorption at the channel layer top surface is proposed to explain why an unpassivated TFT exhibits significantly more NBIS than a passivated TFT.     

Subroutine based detection of APT malware

Statistical detection of mass malware has been shown to be highly successful. However, this type of malware is less interesting to cyber security officers of larger organizations, who are more concerned with detecting malware indicative of a targeted attack. Here we investigate the potential of statistically based approaches to detect such malware using a malware family associated with a large number of targeted network intrusions. Our approach is complementary to the bulk of statistical based malware classifiers, which are typically based on measures of overall similarity between executable files. One problem with this approach is that a malicious executable that shares some, but limited, functionality with known malware is likely to be misclassified as benign. Here a new approach to malware classification is introduced that classifies programs based on their similarity with known malware subroutines. It is illustrated that malware and benign programs can share a substantial amount of code, implying that classification should be based on malicious subroutines that occur infrequently, or not at all in benign programs. Various approaches to accomplishing this task are investigated, and a particularly simple approach appears the most effective. This approach simply computes the fraction of subroutines of a program that are similar to malware subroutines whose likes have not been found in a larger benign set. If this fraction exceeds around 1.5 %, the corresponding program can be classified as malicious at a 1 in 1000 false alarm rate. It is further shown that combining a local and overall similarity based approach can lead to considerably better prediction due to the relatively low correlation of their predictions.     

Towards high accuracy calibration of electron backscatter diffraction systems

For precise orientation and strain measurements, advanced Electron Backscatter Diffraction (EBSD) techniques require both accurate calibration and reproducible measurement of the system geometry. In many cases the pattern centre (PC) needs to be determined to sub-pixel accuracy. The mechanical insertion/retraction, through the Scanning Electron Microscope (SEM) chamber wall, of the electron sensitive part of modern EBSD detectors also causes alignment and positioning problems and requires frequent monitoring of the PC. Optical alignment and lens distortion issues within the scintillator, lens and charge-coupled device (CCD) camera combination of an EBSD detector need accurate measurement for each individual EBSD system.This paper highlights and quantifies these issues and demonstrates the determination of the pattern centre using a novel shadow-casting technique with a precision of ∼10 μm or ∼1/3 CCD pixel.