交互外形设计中的有理插值方法

在CAGD中,构造光滑曲线通过给定的二维或三维点集是非常重要的问题。广泛釆用的方法是Hermite技术,但是依照这种方法,插值曲线的形状过多地依赖于切矢的模长,且形状的修改不能依几何的方式交互完成。本文提出了一种结合交互技术和插值技术于一体的曲线插值方法,插值基函数为有理函数,在此基础上讨论了形状参数在几何修形中的作用,     

有理Bézier曲线的权因子计算

给出了有理Bézier曲线形状作定量修改时的权因子的计算公式,并且证明了权因子选取的唯一性。     

保形光顺的样条曲线

给出了一种对给定平面上的数据点进行保形的最小二乘逼近的算法,并且得到的逼近曲线是光顺的。本文利用拉格朗日乘数法对带约束条件的参数曲线进行了优化设计。     

涡轮平面叶栅非轴对称端壁优化设计

开发了一套造型灵活直观、网格生成速度快的涡轮平面叶栅非轴对称端壁优化设计工具,该工具的核心技术是非均匀有理B样条(NURBS)曲面造型和网格变形.在此基础上以商业软件Isight为优化驱动器,以CFX为求解器,搭建了非轴对称端壁优化设计流程.以Pack B涡轮平面叶栅为例,对其进行了非轴对称端壁优化设计.优化后涡轮平面叶栅总压损失系数减小了12.96%.结果表明:涡轮平面叶栅端部的静压分布改变削弱了涡轮平面叶栅通道中马蹄涡、通道涡的强度,提高了涡轮平面叶栅的气动性能.      

B-样条曲线奇点的几个定理

根据B-样条曲线控制多边形的剖分理论和B-样条曲线及其基函数的性质,分别给出B-样条曲线出现奇点的两个充分必要条件。     

非均匀横流作用下冲击射流冷却的数值研究

运用数值计算的方法对非均匀横流作用下冲击射流冷却的流动和传热过程进行了三维数值研究,揭示出射流冲击孔和横流孔相对排列方式、射流雷诺数和横流雷诺数相对大小、射流孔中心至横流孔出口截面距离等因素对冲击射流冷却流动传热特性的影响。结果表明:若保持总的冷却流量不变,当横流雷诺数与射流雷诺数之比等于或大于0.5后,有横流时的冲击射流冷却局部努塞尔数比无横流时有较为显著的下降;在非均匀的横流作用下,对于叉排和顺排方式,由于射流和横流的相互作用,壁面温度分布呈现出复杂的特征;射流孔和横流孔呈顺排布置时,峰值努塞尔数所对应的射流驻点区向下游偏移。      

B样条快速求值算法及其在曲线拟合中的应用

B样条方法是数据插值、拟合与平滑的重要方法。通过对de Boor-Cox算法计算路径的分析,给出了一种基于向量扩展的B样条基函数求值运算方法。分析表明,该方法具有并行计算结构和递推计算结构,利于计算机实现。同de Boor-Cox递推运算方法比,它能够同步计算出k次B样条的所有k＋1个非零函数值,运算效率提高了2k＋1倍。在最小二乘法B样条曲线拟合算法中的应用验证了本文算法的有效性。     

变厚度弹性件柔性接头的非线性分析

结合试验数据,采用Mooney-Rivlin模型模拟了柔性接头的弹性件材料;根据非线性理论,采用混合插值有限元法对某一摆角状态下的柔性接头进行了大变形非线性弹性分析,并在此基础上初步探讨了厚度不均匀弹性件对层间应力的影响。计算结果表明,弹性层减薄到一定程度时,层间剪切应力沿宽度方向分布变得均匀,应力值仍集中在2 MPa以下;层间法向应力分布变化较小,呈二次曲线分布,最大值有所降低;其中,接近后法兰的弹性件厚度不均时,对各项应力最大值影响较大。     

应用非均匀B样条拟合涡轮发动机部件特性

为了使气动热力模型更好的逼近涡轮发动机,应用成熟的非均匀B样条来拟合部件特性.这种方法既可以保证发动机计算的整体性能也大大提高了其局部特性的准确性.单轴涡喷发动机的算例证明了方法的有效性.      

Rapid local ionosphere modeling based on Precise Point Positioning over Iran: A case study under 2014 solar maximum

Presently, the ionosphere effect is the main source of the error in the Global Positioning System (GPS) observations. This effect can largely be removed by using the two-frequency measurements, while to obtain the reasonable results in the single-frequency applications, an accurate ionosphere model is required. Since the global ionosphere models do not meet our needs everywhere, the local ionosphere models are developed. In this paper, a rapid local ionosphere model over Iran is presented. For this purpose, the GPS observations obtained from 40 GPS stations of the Iranian Permanent GPS Network (IPGN) and 16 other GPS stations around Iran have been used. The observations have been selected under 2014 solar maximum, from the days 058, 107, 188 and 271 of the year 2014 with different geomagnetic activities. Moreover, ionospheric observables based on the precise point positioning (PPP) have been applied to model the ionosphere. To represent our ionosphere model, the B-spline basis functions have been employed and the variance component estimation (VCE) method has been used to regularize the problem.To show the efficiency our PPP-derived local ionosphere model with respect to the International GNSS Service (IGS) global models, these models are applied on the single point positioning using single-frequency observations and their results are compared with the precise coordinates obtained from the double-differenced solution using dual-frequency observations. The results show that the 95th percentile of horizontal and vertical positioning errors of the single-frequency point positioning are about 3.1 and 13.6?m, respectively, when any ionosphere model are not applied. These values significantly improve when the ionosphere models are applied in the solutions. Applying CODE’s Rapid Global ionosphere map (CORG), improvements of 59% and 81% in horizontal and vertical components are observed. These values for the IGS Global ionosphere map (IGSG) are 70% and 82%, respectively. The best results are obtained from our local ionosphere model, where 84% and 87% improvements in horizontal and vertical components are observed. These results confirm the efficiency of our local ionosphere model over Iran with respect to the global models. As a by-product, the Differential Code Biases (DCBs) of the receivers are also estimated. In this line, we found that the intra-day variations of the receiver DCBs could be significant. Therefore, these variations must be taken into account for the precise ionosphere modeling.