表面强化件的疲劳强度分析及金属的内部疲劳极限

本文通过对喷丸，渗氮及碳氮共渗试样疲劳行为的研究，提出了金属内部疲劳极限的概念，指出内部疲劳极限约为表面疲劳极限（一般意义上的疲劳极限）的１．３５倍，并利用这个概念分析了表面强化件的疲劳强度。     

弹簧表层优化喷丸残余应力场的工程计算方法

以往多是依靠疲劳试验或经验来选择弹簧的喷丸强化工艺参数，这种传统作法不仅人力和物力耗费大，而且所确定的工艺参数也不一定具有最佳的强化效果。依据作者提出的并已获得试验验证的“内部疲劳极限理论”，提出一种通过工程计算获得达到预先规定疲劳强度应具备的“优化残余应力场”。这样便有可能把喷丸强化工艺技术在制造业中的应用由以往单纯依靠试验和经验而步入到以工程计算为基础与少量典型试验相结合的更科学的途径。     

高能喷丸过程中的物质转移现象

采用高能喷丸技术在铝合金表面制备出纳米晶结构层，利用金相显微镜（OM），扫描电子显微镜（SEM／EDS），透射电子显微镜（TEM）及X射线衍射仪（XRD）等设备对高能喷丸铝合金表面层的显微结构及成分进行分析。结果表明：在高能喷丸表面纳米化过程中，弹丸中的Fe，cr原子在强制机械力的作用下转移进入铝合金表面，在材料表层约30μm的深度范围内形成分布不均匀的合金化层。这可能是纳米晶体中大量界面的形成使合金元素原子能够快速扩散和偏聚的结果。     

Evaluation of Surface Residual Stresses in Friction Stir Welds Due to Laser and Shot Peening

The effects of laser, and shot peening on the residual stresses in friction stir welds (FSW) has been investigated. The surface residual stresses were measured at five different locations across the weld in order to produce an adequate residual stress profile. The residual stresses before and after sectioning the coupon from the welded plate were also measured, and the effect of coupon size on the residual stress relaxation was determined and characterized. Measurements indicate that residual stresses were not uniform along the welded plate, and large variation in stress magnitude could be exhibited at various locations along the FSW plate. Sectioning resulted in significant residual stress relaxation in the longitudinal direction attributed to the large change in dimensions in this direction. Overall, laser and shot peening resulted in a significant reduction in tensile residual stresses at the surface of the specimens.     

Event-driven,pattern-based methodology for cost-effective development of standardized personal health devices

Experiences applying standards in personal health devices (PHDs) show an inherent trade-off between interoperability and costs (in terms of processing load and development time). Therefore, reducing hardware and software costs as well as time-to-market is crucial for standards adoption. The ISO/IEEE11073 PHD family of standards (also referred to as X73PHD) provides interoperable communication between PHDs and aggregators. Nevertheless, the responsibility of achieving inexpensive implementations of X73PHD in limited resource microcontrollers falls directly on the developer. Hence, the authors previously presented a methodology based on patterns to implement X73-compliant PHDs into devices with low-voltage low-power constraints. That version was based on multitasking, which required additional features and resources. This paper therefore presents an event-driven evolution of the patterns-based methodology for cost-effective development of standardized PHDs. The results of comparing between the two versions showed that the mean values of decrease in memory consumption and cycles of latency are 11.59% and 45.95%, respectively. In addition, several enhancements in terms of cost-effectiveness and development time can be derived from the new version of the methodology. Therefore, the new approach could help in producing cost-effective X73-compliant PHDs, which in turn could foster the adoption of standards.     

基于SDNA-GA 优化的模糊神经网络控制

针对最新的生物DNA研究,病毒中同一DNA碱基顺序可以编码出2条或者3条不同的多肽链.在此基础上分析与模仿了重叠基因和重叠密码的机理,得到一种新的基于重叠基因编码框架,从而提高了问题求解的效率;同时,得到一种移码解读框架的DNA遗传算法(SDNA-GA)计算模型,并将其应用于一类广义隶属度型T-S模糊神经网络控制器(GTS-FNNC)的优化设计,实现了GTS-FNNC的在线学习.     

基于H.264编码域的显著运动检测算法

视频监控中运动目标检测算法易受背景、噪声等因素的影响,导致运动检测准确性降低。针对该问题,提出一种基于H.264编码域的显著运动目标检测改进算法。该算法基于编码过程中帧间运动预测模块输出像素值的绝对差值和,计算其标准差以作为运动判据,同时自适应设置运动判断阈值,以降低背景和噪声的影响。实验结果表明,相比于传统的低复杂度算法,该算法能有效降低背景和噪声等因素的影响,提高显著运动目标的检测准确性。相比于传统的运动检测算法,该算法保持了计算复杂度低的特点,算法复杂度仅为帧间差分法的18.3%,更适用于实时性要求高的监控系统。     

频域二元纯相位编码压缩成像

超分辨率被认为是光学成像和图像处理的"圣杯"之一.压缩感知理论的引入给出一种新的实现从单幅低分辨率图像的超分辨率重构方法,避免了传统超分辨率方法需要多幅亚像素图像的弊端.在分析压缩感知测量矩阵与光学成像系统之间对应约束异同的基础上,提出一种基于4-f光学架构的频域二元相位编码压缩成像方法,可以实现在单次曝光条件下获得的单幅低分辨率测量图像中实现超分辨率重建,不需要其他任何额外的信息采集.二元相位掩膜比均匀相位掩膜更容易物理实现,是压缩成像物理实现的一种更加可行的方案.模拟实验表明提出的方法可以有效地捕获图像的信息与高精度重建.此外,对于大尺度图像重建,该方法在重建时间方面优于Romberg提出的随机解调方法,在更符合实际方面的采样策略方面优于Yin提出的RecPC方法.     

基于字频分布的中文网页编码识别算法

编码识别是网页内容过滤的必要前提,多种中文编码共存给中文网页的内容过滤带来不便。针对上述问题,提出一种基于字频分布的中文网页编码识别算法。根据汉字的使用频率,选取使用频度较高的字符构成高频字符编码表,以高频字符编码作为关键字,使用改进的模式匹配算法查找待识别网页,并统计匹配次数。将编码的匹配结果作为分析的依据,最终判定待识别网页的真实码制。实验结果证明,与Unigram算法相比,该算法对目前通用的中文编码识别率较高,适合对未知编码的中文网页进行快速编码识别。     

一种改进的MH编码算法—MHZ算法

针对MH编码算法的不足做了改进,提出了一种新的编码算法——MHZ编码算法.传统的MH编码算法只考虑了图像水平方向上像素之间的相关性,而MHZ编码算法不仅考虑了水平方向上像素之间的相关性,还考虑了垂直方向上像素之间的相关性,在进行编码时,选择垂直高度为2像素的一组数据同时进行编码,使得垂直方向上具有相关性的像素可以一次性地进行编码,从而减少图像的编码数据,进一步提高数据的压缩率.论文最后对五种常用的传真数据分别采用MHZ和MH的编码方式就压缩率进行了对比分析.实验表明MHZ编码算法的效率比传统MH编码算法提高了4％～39％.