SIMULATION OF COMPOSITE NON-LINEAR MECHANICAL BEHAVIOR OF CMCS BY FEM-BASED MULTI- SCALE APPROACH

The non-linear behavior of continuous fiber reinforced C/SiC ceramic matrix composites （CMCs） under tensile loading is modeled by three-dimensional representative volume element （RVE） models of the composite. The theoretical background of the multi-scale approach solved by the finite element method （FEM） is recalled first- ly. Then the geometric characters of three kinds of damage mechanisms, i.e. micro matrix cracks, fiber/matrix interface debonding and fiber fracture, are studied. Three kinds of RVE are proposed to model the microstructure of C/SiC with above damage mechanisms respectively. The matrix cracking is modeled by critical matrix strain en- ergy （CMSE） principle while a maximum shear stress criterion is used for modeling fiber/matrix interface debond- ing. The behavior of fiber fracture is modeled by the famous Weibull statistic theory. A numerical example of con- tinuous fiber reinforced C/SiC composite under tensile loading is performed. The results show that the stress/ strain curve predicted by the developed model agrees with experimental data.     

大变形异步叠轧技术制备高强高导超细晶铜材研究

采用大变形异步叠轧并结合退火处理制备出了超细晶铜材。讨论了大变形异步叠轧的搓轧区、界面以及变形量等技术特征,测试了超细晶铜材的组织及性能。结果表明,搓轧区的存在促进了界面的复合和晶粒的细化;纯铜经过室温六道次AARB变形(累积真应变4.01)后,包含有许多亚结构,之后再经过220℃/35min退火处理,亚结构消失,获得了平均晶粒大小为200nm、抗拉强度、屈服强度分别为424.5MPa,323.1MPa、电导率为76.3 MS.m-1的高强高导超细晶铜;晶界面积的增加提高了超细晶铜的显微硬度;超细晶粒的变形不均匀性小和应力集中小,使得超细晶铜的塑性好、强度高;超细晶铜材的塑性变形机制受晶界行为控制。     

基于区域划分与标准时间的手部异常行为检测

通过监控视频自动检测操作任务中手部异常行为，能够预防人因差错，提高人因可靠性。针对手部操作任务运动特征不明显、常用异常检测和手势识别方法不适用的问题，提出基于区域划分与标准时间的手部异常行为检测技术。使用基于椭圆模型的肤色检测方法，检测视频中手部形心位置；提出工作区域划分方法，根据手部形心在操作过程中所处区域的不同，将连续操作分割为单元任务，获得各段单元任务的起止时间和持续时长；以正常工作时间为标准，对超出标准时间范围的单元任务提出异常警告。实验表明:所提方法的单元任务分割正确率高于93%，异常行为检测率高于86%，能够有效检测手部异常行为，为人为差错的监测与预警提供技术支持。