甚高频数据链模式二协议栈的软件框架设计研究

为解决协议工作效率低下的问题,参考HDLC协议软件的设计方法,本文提出了甚高频数据链模式二(VDL M2)协议的软件模型和设计思想。首先,介绍了VDL M2协议栈的层次架构,并简要阐述了各层的功能。然后,介绍了协议栈的软件设计与实现方法,在不改变Linux内核现有地址族的前提下,创建基于本协议的新套接字地址族AF_VDLM2的操作。依据功能操作,软件结构确立为模块化结构布局;着重分析了协议栈数据收发模块的软件设计流程及缓冲区的结构设计。最后,介绍了协议层之间的通信方式——共享内存区的数据结构及创建和操作。用该方法设计的协议栈可方便地加载到机载通信单元,大大节省了设备成本。     

ALGEBRAIC TURBULENCE MODEL WITH MEMORY FOR COMPUTATION OF 3-D TURBULENT BOUNDARY LAYERS WITH VALIDATION

Additional equations were found based on experiments for an algebraic turbulence model to improve the prediction of the behavior of three dimensional turbulent boundary layers by taking account of the effects of pressure gradient and the historical variation of eddy viscosity, so the model is with memory. Numerical calculation by solving boundary layer equations was carried.out for the five pressure driven three dimensional turbulent boundary layers developed on flat plates, swept-wing, and prolate spheroid in symmetrical plane. Comparing the computational results with the experimental data, it is obvious that the prediction will be more accurate if the proposed closure equations are used, especially for the turbulent shear stresses.     

C++Builder语言的虚拟设备驱动程序VxD的开发

本文介绍了虚拟设备驱动程序的基本概念和用C Builder开发VxD程序的方法，并以在证券业务屏中普遍采用内存映射卡为例进行简略说明。     

一种改进的变记忆长度辨识算法

本文基于多分辨分析方法 ,提出了一种改进的变记忆长度的小波变换多分辨分析方法 ,该方法将辨识的数据分解到不同频段 ,然后针对各不同频段采用不同的记忆长度进行辨识 ,通过对某航空涡轮起动机系统辨识的实例 ,表明该方法可以有效地提高系统辨识的精度。     

EEA/DCP材料的形状记忆特性研究

对以过氧化二异丙苯(DCP)为交联剂的乙烯-丙烯酸乙酯共聚物(EEA)体系进行了研究,结果发现DCP的用量对形状回复率、形状固定率、回复响应温度、拉伸强度和断裂延伸率有很大的影响,并由此确定了当DCP用量为0.5%时,EEA体系具有优良的形状记忆性能;此外对材料动态性能的测试,指出了随着DCP用量的增加,室温模量逐步降低,而高温模量逐步提高,该实验同时还揭示了形状记忆高分子的高弹平台特性.      

二元智能复合材料结构变形分析

提出了嵌入形状记忆合金丝和压电陶瓷片作为联合变形驱动器的二元智能结构,并建立了有限元模型．驱动器镶嵌于层板表面或内部,其中压电陶瓷片作为压电铺层;形状记忆合金丝的驱动力当量为与温度相关的节点载荷．算例比较了不同种驱动方式的变形效果,并进行了试验验证．结果表明在二元智能结构的形状控制中,形状记忆合金丝承受主要的载荷,压电陶瓷片作变形修正,可以发挥各自的优点．      

Integrated optimization design of smart morphing wing for accurate shape control

Smart morphing wing, which is equipped with smart materials and able to change structural geometry adaptively, can further improve aerodynamic efficiency of aircraft. This paper presents a new integrated layout and topology optimization design for morphing wing driven by shape memory alloys (SMAs). By simultaneously optimizing the layout of smart actuators and topology of wing substrate, the ultimately determined configuration can achieve smooth, continuous and accurate geometric shape changes. In addition, aerodynamic analysis is carried out to compare smart morphing wing with traditional hinged airfoil. Finally, the optimized smart wing structure is constructed and tested to demonstrate and verify the morphing functionality. Application setbacks are also pointed out for further investigation.     

Modeling bending behavior of shape memory alloy wire-reinforced composites: Semi-analytical model and finite element analysis

In this study, we propose a novel and simple exact semi-analytical model for superelastic Shape Memory Alloy (SMA) wire reinforced composites subjected to bending loads. In order to study the mechanical response of the composite during loading/unloading, a Representative Volume Element (RVE) is extracted to examine the bending response of the composite. Analytical moment–curvature, and shear force-shear strain relations are derived based on a 3-Dimensional (3D) thermomechanical SMA model and Timoshenko beam theory. The composite Simpson’s rule is adopted to numerically solve the exact analytical moment–curvature and shear force-shear strain relationships. Results including the moment–curvature response, axial stress distribution along the vertical and longitudinal directions, martensite volume fraction, and the tip deflection are reported and validated against 3D finite element simulations. The influence of temperature, martensite volume fraction distribution, and matrix stiffness on the mechanical performance of the composite is also investigated. In particular, the composite is found to behave superelastically under certain conditions of temperature, SMA volume fraction, and elastic stiffness of the matrix. Such behavior is advantageous in applications requiring large recoverable strains or high energy dissipation density.     

多层自适应模式识别系统模型

在英国WISARD单层模式识别系统的基础上,借助P.Kanerva的稀疏分布存贮(SDM)的概念,提出了~种新的多层自适应模式识别系统模型(MAPR)。并就其工作过程和主要特点作了较详细的叙述,还列出了多体印刷体汉字识别的初步试验结果。MAPR用稀疏RAN阵列代替WISARD的常规RAM阵列,用对n元模式计频的训练策略代替了原系统的直接置位策略。使系统除了保持原系统的重要优点外,在大维数或非确定性模式数据识别方面,其性能有了明显改善。     

形状记忆合金增强复合材料的一维σ－ε－T关系

分形状记忆合金（ＳＭＡ）相变前和相变过程中两阶段，推导了形状记忆合金增强复合材料（ＳＭＡＲＣ）的一维应力、应变和温度的关系。实验研究包括：ＮｉＴｉｎｏｌＳＭＡ丝自由回复时的回复应变与温度、受限回复下的回复应力与温度的关系，玻璃纤维／环氧树脂复合材料和ＳＭＡＲＣ试件的单向拉伸，一定应变下ＳＭＡＲＣ拉伸试件的应力与温度的关系，恒定外载下ＳＭＡＲＣ拉伸试件和纯弯曲试件在触发ＳＭＡ前后应变的变化。结果表明，ＳＭＡＲＣ除了具备传统结构材料承载的功能之外，还具有改善和增强材料本身性能的能力。