矢量喷管数学模型研究

采用计算流体力学的有限体积法和神经网络方法,对矢量喷管的统一数学模型进行了研究,并得到了一种以神经网络描述形式的数学模型。该模型为矢量喷管控制规律和带有推力矢量的飞行／推进系统综合控制的研究提供了一定基础。     

疲劳试验中的数据处理

从疲劳试验工作的实际出发，介绍与疲劳问题有关的数理统计方法。该方法使分散性很大的疲劳试验结果得以合理处理，从而可准确地提供试验数据，供作材料疲劳鉴定与疲劳寿命估算之用。     

样条有限层法在复合材料板和圆柱壳中的应用

由于样条函数对各种边界条件的适用性和平板曲板通用性,本文应用三次B月样条有限层法分析复合材料板壳,推导出适用于金属、复合材料板和圆柱壳单元的通用公式,并通过算例说明使用该模型的优越性。     

一种修正的次优子集多项式回归法

提出以集合论为工具建立在抽象代数理论基础上的修正次优子集多项式回归法，对回归过程的主要内容进行了充分的数学描述和讨论，给出了一个重要的定理和推论。理论分析和实际计算发现，利用本文提出的修正次优子集回归法对给定的任意精度都可求出最佳逼近结果。     

空间对接动力学仿真软件DODASS及其工程应用

首先简要阐述了研究空间对接动力不的目的意义，主要技术难点和攻关内容及其数学模型建立的方法和步骤；然后重点介绍了基于周边式对接机构系统的空间对接动力学模型而开发成功的ＤＯＤＡＳＳ分析仿真软件及其工程应用的初步仿真曲线；最后给出了可供工程设计参考的初步结论。     

深孔尺寸及形状误差自动综合测量理论与方法的研究

通过研究深孔尺寸及其形状误差在线自动综合测量的基本理论,给出一种新的误差分离方法。用此方法,能够在测量过程中将被测深孔工件尺寸及形状误差与工件的回转运动误差、测头的直线运动误差分离开来。而且在进行误差分离的同时,能精确确定被测表面各点的三维坐标,建立起各项被测参数的数学模型,并研制成功由微机实时数据处理的深孔在线综合测量系统。实验结果表明,该测量系统的基本理论正确,测量结果准确可靠。     

三波长辐射温度计

采用辐射法测量物体表面真实温度的最大困难是被测材料表面的发射率难测。三波长辐射温度计能消除发射率对测温的影响,尤其是对抛光表面的测温。本文基于推导多波长辐射温度计的理论,推出了一种三波长辐射测温法的新关系式,并建立了三波长辐射温度计的数学模型。     

对 k-ω 湍流模型的改进

为了更适合于较复杂几何外形的数值计算，本文提出仅引入１个参数来改进ｋ－ω湍流模型并采用了一种双网格系统的有限体积格式来求解Ｎ－Ｓ方程与改进的的ｋ－ω模型方程。通过平板附面层的计算结果表明了用本文提出的对ｋ－ω湍流模型的改进对控制转捩是有效的。     

环形散热器的动态性能

 利用热阻热容概念近似描述环形散热器微元段上的动态特性,加上热平衡条件,从而推导出整个环形散热器动态数学模型。只要在微元段上选取适当的节点数,该模型就可以满足任意精度之要求。借助于计算机并使用有限差分法,对环形散热器的动态性能进行了数值计算,理论计算结果与空中飞行试验数据吻合较好。还给出了环形散热器芯体在某一状态下的温度分布,作为动态响应计算的特例,最后计算了此状态下环形散热器稳态性能,其结果与对应的试飞数据比较令人非常满意。     

Applying the mathematical theory of stochastic processes to lunar and planetary science: the ancient oceans on Mars case

The Mathematical Statistics Theory (MST) and the Mathematical Theory of Stochastic Processes (MTSP) are different branches of the more general Mathematical Probability Theory (MPT) that represents different aspects of some physical processes we can analyze using mathematics. Each model of a stochastic process according to the MTSP can provide one or more interpretations in the MST domain. The importance of MTSP is that each such interpretation can provide large amount of new information. While large body of work on the impact crater statistics according to MST has already been done, it is yet to be investigated as to how we can model a stochastic process according to MTSP; for example, bombardment of the planetary surface or something else in a Lunar and Planetary Science (LPS) domain. In order to show possible achievements, the possible existence of a Martian ocean was chosen as a query that could be addressed through computations using presumptions according to MTSP, including probability of existence as well as lateral and vertical extent and duration of time. While the presumptions for this particular case will also be addressed in certain degree, this will be done primarily to show complexities of some physical process that can be modeled, rather than to prove correctness of the concrete values computed here. While this in itself can be the objective in some future work toward the formal proof of the probability, extent, and timing of oceans on Mars, the basic idea here is to show the basic principles of MTSP, and its potential for addressing LPS-related phenomena. Here, I attempt to show how this approach can: (1) provide large amounts of previously unknown information about physical processes on the surface of the planet, (2) lead to a better understanding of the processes that have shaped the surface of the planet, and/or (3) help constrain the amount of resurfacing. Coupled with other current methodologies, MTSP can result in a better understanding of the history of Mars, as well as other lunar and planetary bodies.