纵向数据非参数模型的惩罚修正二次推断函数估计

基于纵向数据研究非参数模型y=f(t)+ε,其中f(·)为未知平滑函数,ε为零均值随机误差项.利用截断幂函数基对f(·)进行基函数展开近似,并且结合惩罚样条的方法构造关于基函数系数的惩罚修正二次推断函数.然后利用割线法迭代得到基函数系数估计的数值解,从而得到未知平滑函数的估计.理论证明,应用此方法所得到的基函数系数估计具有相合性和渐近正态性.最后通过数值方法得到了较好的拟合结果.     

基于三次Bézier基函数插值的GM(1,1)模型背景值优化研究

本文研究了传统灰色GM(1,1)模型存在模型精度不高的问题.利用带形状参数的三次Bézier基函数,给出插值函数的表达式,并结合复化梯形公式,给定误差限的方法,获得了比传统灰色GM(1,1)模型更高精度的结果.推广了传统灰色GM(1,1)预测模型的结果.     

基于RLS算法的多项式预测模型及其应用研究

为了提高经济领域统计数据的预测精度,代数多项式预测模型的建模方法应运而生.该方法使用代数多项式模型拟合给定的经济统计数据,并使用递推最小二乘法(RLS)对多项式拟合模型的加权系数进行递推计算以获得最优模型参数,然后通过获得的最优多项式模型计算未来预测数据.文章以实际统计的经济数据为例进行了仿真计算,研究结果表明,该方法不仅能实现统计数据的高精度拟合,而且具有很好的预测能力,在经济领域具有广阔的应用前景.     

参数化设计中确定参数有效范围的吴特征列方法

参数化CAD设计中,需要对给定的草图进行修改进而得到满足设计者需求的模型.然而,在修改参数值时,常常由于给定的参数值不合理,而导致无法重新生成几何图形.利用吴特征列方法从代数可解的角度给出了一个在参数化设计中确定参数的有效范围的算法.实例的分析计算,证明算法是可行的.     

一种新的二次约束二次规划问题的分支定界算法

本文为了获得二次约束二次规划(QCQP)问题的全局最优解,提出一种新的参数化线性松弛分支定界算法.该算法利用参数化线性松弛技术,得到(QCQP)的全局最小值的下界,并利用区域缩减技术以最大限度地删除不可行区域,加快该算法的收敛速度.数值实验表明,本文提出的算法是有效并且可行的.     

半变系数模型改进的轮廓最小二乘估计

针对半变系数模型,在局部线性拟合轮廓最小二乘估计方法的基础上将关于变系数函数的局部线性拟合改进为局部非线性拟合,得到半变系数模型改进的轮廓最小二乘估计,进一步讨论了常值系数的渐进正态性.     

基于变体积约束的阻尼材料微结构拓扑优化研究北大核心CSCD

阻尼复合结构的抑振性能取决于材料布局和阻尼材料特性.该文提出了一种变体积约束的阻尼材料微结构拓扑优化方法,旨在以最小的材料用量获得具有期望性能的阻尼材料微结构.基于均匀化方法,建立阻尼材料三维微结构有限元模型,得到阻尼材料的等效弹性矩阵.逆用Hashin-Shtrikman界限理论,估计对应于期望等效模量的阻尼材料体积分数限,并构建阻尼材料体积约束限的移动准则.将获得阻尼材料微结构期望性能的优化问题转化为体积约束下最大化等效模量的优化问题,建立阻尼材料微结构的拓扑优化模型.利用优化准则法更新设计变量,实现最小材料用量下的阻尼材料微结构最优拓扑设计.通过典型数值算例验证了该方法的可行性和有效性,并讨论了初始微构型、网格依赖性和弹性模量等对阻尼材料微结构的影响.     

关于Rudin—Osher—Fatemi图像恢复模型特性的一个注记

本文分析了ROF(Rudin-Osher-Fatemi)模型并给出了它的几何特性.根据不同的目的,对ROF模型来说,可以完全从给定的观察图像来选择特殊的参数λ的值.     

一类新的极小谱任意符号模式

若给定任意一个$n$次首一实系数多项式$f(\lambda)$,都存在一个实矩阵$B\in Q(A)$, 使得$B$的特征多项式为$f(\lambda)$,则称$A$为谱任意符号模式. 如果一个谱任意符号模式的任意非零元被零取代后所得到的符号模式不是谱任意,那么这个谱任意符号模式称为极小谱任意符号模式.本文证明一类极小谱任意符号模式.     

基于参数化水平集法的材料非线性子结构拓扑优化

针对利用传统水平集法进行非线性结构拓扑优化计算过程复杂及计算效率低等问题,将参数化水平集方法引入材料非线性结构拓扑优化中。通过全局径向基函数插值初始水平集函数,建立了以插值系数为设计变量、结构的应变能最小为目标函数、材料用量为约束条件的材料非线性结构拓扑优化模型,利用有限元分析对材料非线性结构建立平衡方程,并用迭代法求解。同时,采用子结构法划分设计区域为若干个子区域,将全自由度平衡方程的求解分解为缩减的平衡方程和多个子结构内部位移的求解,减小了计算成本。算例表明,这种处理非线性关系的方法可以在保证数值稳定的同时提高计算效率,得到边界清晰、结构合理的拓扑优化构形。     

预测水驱油藏中油、水流动问题的快速方法

以最佳正交分解(POD)技术为基础提出了一种快速预测油藏中油、水流动问题的方法.采用POD技术建立了水驱油藏中油、水两相流动的低阶模型.通过油藏数值模拟方法获得二维水驱油藏模型在时间0～500 d内的压力和含水饱和度的100个样本, 并从样本中提取出一组压力和含水饱和度的POD基函数.当注采参数不断变化后,采用已求得的POD基函数结合低阶模型对新的物理场进行预测.研究结果表明:POD方法能够快速、准确地预测出水驱油藏的压力和含水饱和度场,文中算例给出压力和含水饱和度场的预测误差分别不超过1.2%与1.5%,且计算速度比直接进行油藏数值模拟快50倍以上.     

A Prediction Model for Skin Wound Suture Forces With Uncertain Material Parameters北大核心CSCD

To evaluate the forces required for the suture of skin wounds quickly and effectively, the nonlinear finite element method was used to calculate the suture forces for skin wounds with different sizes and material parameters. With the calculated results as samples, the prediction model for skin wound suture forces was constructed by means of the EBF neural network model. Given the uncertain skin material parameters influencing the reliability of numerical results, the Monte-Carlo method was used to analyze the uncertainty propagation of skin material parameters. Finally, the prediction analysis and measuring experiment of wound suture forces were carried out with pig skin specimens to verify the reliability of the method. The results showed that, the suture force increases first and then decreases according to the suture point sequence, and the peak force occurs before the center of the wound. For a 40 mm×10 mm wound, the peak suture force is about 1.7 N, and that for a 40 mm×14 mm wound is about 2.5 N. Influenced by the uncertainty of material parameters, the prediction results of suture forces fluctuate by as much as ±0.6 N. The proposed theoretical prediction model provides an effective solution to the problem of parameter uncertainty propagation for biological soft tissue materials such as skins, and makes an important mechanical reference for robotic surgical suture. © 2023 Editorial Office of Applied Mathematics and Mechanics. All rights reserved.     

Learning patient-specific parameters for a diffuse interface glioblastoma model from neuroimaging data

Parameters in mathematical models for glioblastoma multiforme (GBM) tumour growth are highly patient specific. Here, we aim to estimate parameters in a Cahn–Hilliard type diffuse interface model in an optimised way using model order reduction (MOR) based on proper orthogonal decomposition (POD). Based on snapshots derived from finite element simulations for the full-order model (FOM), we use POD for dimension reduction and solve the parameter estimation for the reduced-order model (ROM). Neuroimaging data are used to define the highly inhomogeneous diffusion tensors as well as to define a target functional in a patient-specific manner. The ROM heavily relies on the discrete empirical interpolation method, which has to be appropriately adapted in order to deal with the highly nonlinear and degenerate parabolic partial differential equations. A feature of the approach is that we iterate between full order solvers with new parameters to compute a POD basis function and sensitivity-based parameter estimation for the ROM problems. The algorithm is applied using neuroimaging data for two clinical test cases, and we can demonstrate that the reduced-order approach drastically decreases the computational effort.     

Identification of the Material Parameters of a Micropolar Model for Granular Materials

Granular frictional materials show a complex stress‐strain behaviour depending on the stress state and the load history. Furthermore, biaxial experiments exhibit the occurrence of shear band phenomena as the result of the localization of plastic strains. It is well known that the onset of shear bands is associated with microrotations of the granular microstructure, which has a significant influence on the macroscopic behaviour. Consequently, the macroscopic material must result in a micropolar model, which incorporates rotational degrees of freedom. After the formulation of the constitutive equations and the numerical implementation, it is necessary to determine all required material parameters. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Model reduction of porous-media problems using proper orthogonal decomposition

In the context of finite-element simulations of porous media, computing time and numerical effort is an important issue because the number of degrees of freedom of such coupled problems can become very large. Following this, model reduction plays an important role. A broad variety of materials exhibit a porous microstructure. In order to evaluate the overall response of these materials, a macroscopic continuum-mechanical modelling approach is used. Therefore, the complex inner structure of porous media is regarded in a multi-phasic and multi-component manner by means of the well-founded Theory of Porous Media (TPM). The mechanical behaviour of porous media is solved using the Finite-Element Method (FEM). The basic idea of model reduction is to transform a high dimensional system, in terms of the system's degrees of freedom, to a low dimensional subspace to minimise the computational effort while maintaining the accuracy of the solution. The method of proper orthogonal decomposition (POD) can be seen as a method to approximate a given data set with a low dimensional subspace. Furthermore, the POD method is independent of the type of the model and can be used for nonlinear systems as well as for systems of second order. In several applications, such as consolidation problems of partially saturated soils, commonly occurring motion sequences can be found, which can be used as typical “snapshots” of the system. Therefore, the application of the POD method to the simulation of porous media is discussed in the present contribution. Investigated computations of a biphasic standard problem show that the POD method reduces the numerical effort to solve the linearised system of equations in each iteration step. (© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Simulation Based Determination of Piezoelectric Material Parameters

The exact numerical simulation of piezoelectric transducers needs the knowledge of all material tensors that occur in the piezoelectric constitutive relations. The determination of these tensors is achieved by a simulation based algorithm which adjusts the 3D - FEM simulated data with electrical measurements of a piezoelectric transducer. Its advantage compared to the standards (see [1], [2]) lies in the fact that a determination of the complete set of material parameters from one arbitrarily shaped specimen with a high precision is possible. The reconstruction of the material tensors is formulated as a parameter identification problem for a system of PDEs. Since unique solvability of this inverse problem may hardly be verified, the system of equations we have to solve for recovering the material tensor entries can be rank deficient and therefore requires application of appropriate regularization strategies. For this purpose, we use inexact Newton methods. The material parameters are assumed to be complex-valued which allows to account for mechanical, dielectric and piezoelectric losses. (© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

交通流模型基于特征投影分解技术的外推降维有限差分格式

利用特征正交分解(proper orthogonal decomposition,简记为POD)技术研究交通流的Aw-Rascle-Zhang(ARZ)模型. 建立一种基于 POD方法维数较低的外推降维有限差分格式, 并用数值例子检验数值计算结果与理论结果相吻合, 进一步表明基于POD方法的外推降维有限差分格式对于求解交通流方程数值解是可行和有效的.     

Context-specific independence in graphical log-linear models

Log-linear models are the popular workhorses of analyzing contingency tables. A log-linear parameterization of an interaction model can be more expressive than a direct parameterization based on probabilities, leading to a powerful way of defining restrictions derived from marginal, conditional and context-specific independence. However, parameter estimation is often simpler under a direct parameterization, provided that the model enjoys certain decomposability properties. Here we introduce a cyclical projection algorithm for obtaining maximum likelihood estimates of log-linear parameters under an arbitrary context-specific graphical log-linear model, which needs not satisfy criteria of decomposability. We illustrate that lifting the restriction of decomposability makes the models more expressive, such that additional context-specific independencies embedded in real data can be identified. It is also shown how a context-specific graphical model can correspond to a non-hierarchical log-linear parameterization with a concise interpretation. This observation can pave way to further development of non-hierarchical log-linear models, which have been largely neglected due to their believed lack of interpretability.     

Spatial Random Material Property Model for Vibration of Composites

Investigation of vibration and buckling of thin walled composite structures is very sensitive to parameters like uncertain material properties and thickness imperfections. Because of the manufacturing process and others, thin walled composite and other structures show uncertainties in material properties, and other parameters which cannot be reduced by refined discretization. These parameters are mostly spatial distributed in nature. Here I introduce a semivariogram type material property model to predict the spatial distributed material property (like young's modulus) over the structure. The computation of semivariogram parameters needs the local material properties over a prespecified gird. The material properties at each grid have been obtained by considering a statistically homogeneous representative volume element (RVE) at each gird. According to random nature of the spatial arrangement of fibers, the statistically homogeneous RVE is obtained using image processing. The effective material properties of the RVE have been obtained numerically with the help of periodic boundary condition. The methodology is applied to a composite panel model and modal analysis has been carried. The results of the modal analysis (eigen values and mode shapes) are compared with experimental modal analysis results which are in good agreement. Using the presented material property model we can better predict the vibration characteristics of the thin walled composite structures with the inherent uncertainties. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)