Stochastic finite element method based on local averages of random fields

The stochastic finite element method (SFEM) based on the local, averages of random fields, which was proposed in [5], is now generalized to analyze the structures with several correlated random parameters. The covariance matrix of the local averages of a random vector field is derived. The SFEM based on the local averages of random vector fields is formulated. The numerical examples show that the generalized SFEM preserves the advantages of the original one, i. e., rapid convergence, good accuracy and insensitivity to the correlation structures of random parameters.Project supported by National Natural Science Foundation of China.     

精细积分时域平均法和随机扩阶系统法

讨论含随机参数结构的动力响应的计算问题,发展了精细积分时域平均法（ＴＡＰＩＭ）,它可以用来计算确定性系统受到随机激励时的动力响应;结合随机扩阶系统方法与随机有限元法,将ＴＡＰＩＭ方法应用于计算随机参数结构的动力响应,取得了较好的结果。结出了数值算例,结果表明随机扩阶系统法,随机有限元法与精细积分时域平均法的结合是计算 随机参数结构动力响应的有效方法。     

On the local average of random field in stochastic finite element analysis

The distinguishing feature of stochastic finite element analysis is that it involves the discretization of the parameter space of random fields of material properties, the geometry of structure and / or the loads. It is shown in earlier investigations that a reasonable procedure of discretization is to take the local averages of the random fields on each element. In the present paper the formulae for the covariance of the local averages of a homogeneous random vector field on rectangular elements are generalized by relaxing the condition. For an inhomogeneous random field and /or non-rectangular elements, a procedure of using Gaussian quadrature to evaluate the means and covariances of the local averages is proposed. Thus, the stochastic finite element method (SFEM) based on the local averages of random fields is adapted to a structure with irregular shape and / or inhomogeneous random fields. The effects of the mesh geometry, the ratio of element size to the correlation scale as well as the number of Gaussian quadrature points on the convergence of SFEM are discussed. It is found that even better results could be obtained by utilizing appropriate Gaussian quadrature instead of exact local average.Project supported by National Natural Science Foundation of China.     

Equivalent linearization for systems subjected to non-stationary random excitation

The method of eauivalent linearization is applied to the general problem of the response of non-linear discrete systems to non-stationary random excitation. Conditions for minimum equation difference are determined which do not depend explicitly on lime but only on the instantaneous statistics of the response process. Using the equivalent linear parameters, a deterministic non-linear ordinary differential equation for the covariance matrix is derived. An example is given of a damped Duffing oscillator subjected to modulated white noise.     

Theoretical framework and experimental procedure for modelling mesoscopic volume fraction stochastic fluctuations in fiber reinforced composites

Many engineering materials exhibit fluctuations and uncertainties on their macroscopic mechanical properties. This randomness results from random fluctuations observed at a lower scale, especially at the meso-scale where microstructural uncertainties generally occur. In the present paper, we first propose a complete theoretical stochastic framework (that is, a relevant probabilistic model as well as a non-intrusive stochastic solver) in which the volume fraction at the microscale is modelled as a random field whose statistical reduction is performed using a Karhunen–Loeve expansion. Then, an experimental procedure dedicated to the identification of the parameters involved in the probabilistic model is presented and relies on a non-destructive ultrasonic method. The combination of the experimental results with a micromechanical analysis provides realizations of the volume fraction random field. In particular, it is shown that the volume fraction can be modelled by a homogeneous random field whose spatial correlation lengths are determined and may provide conditions on the size of the meso-volumes to be considered.     

The asymptotic stability analysis in stochastic logistic model with Poisson growth coefficient

The asymptotic stability of a discrete logistic model with random growth coefficient is studied in this paper. Firstly, the discrete logistic model with random growth coefficient is built and reduced into its deterministic equivalent system by orthogonal polynomial approximation. Then, the linear stability theory and the Jury criterion of nonlinear deterministic discrete systems are applied to the equivalent one. At last, by mathematical analysis, we find that the parameter interval for asymptotic stability of nontrivial equilibrium in stochastic logistic system gets smaller as the random intensity or statistical parameters of random variable is increased and the random parameter's influence on asymptotic stability in stochastic logistic system becomes prominent.     

A ‘poor man's Navier–Stokes equation’: derivation and numerical experiments—the 2‐D case

We present a systematic derivation of a discrete dynamical system directly from the two‐dimensional incompressible Navier–Stokes equations via a Galerkin procedure and provide a detailed numerical investigation (covering more than 107 cases) of the characteristic behaviours exhibited by the discrete mapping for specified combinations of the four bifurcation parameters. We show that this simple 2‐D algebraic map, which consists of a bilinearly coupled pair of logistic maps, can produce essentially any (temporal) behaviour observed either experimentally or computationally in incompressible Navier–Stokes flows as the bifurcation parameters are varied in pairs over their ranges of stable behaviours. We conclude from this that such discrete dynamical systems deserve consideration as sources of temporal fluctuations in synthetic‐velocity forms of subgrid‐scale models for large‐eddy simulation. Copyright © 2004 John Wiley & Sons, Ltd.     

改进的区间截断法及基于区间分析的非概率可靠性分析方法

从工程应用的角度出发 ,提出了用区间表示参数的不确定性时 ,线性系统的非概率可靠度指标 ,此可靠度指标在试验数据较少时比概率可靠度指标更合理 ;另外本文还提出了计算响应参量变化范围的区间截断法 ,研究了截断参数 t的选取范围 ,算例分析表明 ,当 t取为所有输入参数的最大相对变化量时 ,由区间截断法算得的结果近似等于精确解 ,而随着 t的增大 ,区间截断法的解逐渐平稳地趋近于由区间算术运算所求得的直接解。     

A homogenized Love–Kirchhoff model for out-of-plane loaded random 2D lattices: Application to “quasi-periodic” brickwork panels

A homogenization procedure for finding the bending stiffness of a 2D regular lattice with random local interactions is proposed. The kinematic and static methods are used to provide explicit upper and lower bounds for the homogenized moduli. The proposed homogenization procedure is applied to a masonry obtained by a random perturbation of the periodic running bond masonry [Cecchi, A., Sab, K., 2009. Discrete and continuous models for in plane loaded random elastic brickwork. Eur. J. Mech. A 28, 610–625].A numerical evaluation of the scatter between the discrete models and the 2D Love–Kirchhoff model is performed on a test case, for various values of the random perturbation parameter and of the parameter that characterizes the heterogeneity of the wall. As expected, when the number of heterogeneities in the structure is large enough, the average response of the random discrete model converges to an asymptotic response. It is shown that this asymptotic response is very close to that of the periodic discrete model which is in turn very close to the response of the deterministic homogenized model. Similarly to the conclusion of Cecchi and Sab [Cecchi A., Sab K., 2009. Discrete and continuous models for in plane loaded random elastic brickwork. Eur. J. Mech. A. 28, 610–625.] dedicated to in-plane loading, the present results concerning out-of-plane loading show (both by means of a discrete model and a homogenized model) that the running bond pattern may be used successfully to analyze historical masonries with blocks having irregular widths in the horizontal direction.     

含随机参数的多体系统动力学分析

基于Lagrange方程建立了含随机参数的多体系统的动力学 模型,利用广义坐标分离法将随机微分代数方程转化为随机纯微分方程,利用Newmark法进行数值解算. 应用随机因子法求解系 统随机响应的数字特征,获得统计意义下的解. 以旋转杆滑块系统为例,考虑系统中载荷、物理和几何参数的随机性,通过与Monte Carlo法结果的对比验证了文中方法的正确性和有效性. 计算结果表明,部分随机参数的分散性对多体系统动力响应的影响不可忽略,利用随机参数的动力学模型将能客观地反映出系统的动力学行为.     

Meso-scale modelling of the size effect on the fracture process zone of concrete

The size effect on the fracture process zone in notched and unnotched three point bending tests of concrete beams is analysed by a meso-scale approach. Concrete is modelled at the meso-scale as stiff aggregates embedded in a soft matrix separated by weak interfaces. The mechanical response of the three phases is modelled by a discrete lattice approach. The model parameters were chosen so that the global model response in the form of load-crack mouth opening displacement curves were in agreement with experimental results reported in the literature. The fracture process zone of concrete is determined numerically by evaluating the average of spatial distribution of dissipated energy densities of random meso-scale analyses. The influence of size and boundary conditions on the fracture process zone in concrete is investigated by comparing the results for beams of different sizes and boundary conditions.     

Sensitivity and randomness in homogenization of periodic fiber-reinforced composites via the response function method

The main issue this paper addresses is the derivation and implementation of a general homogenization method, including the simultaneous determination of sensitivity gradients and probabilistic moments of the effective elasticity tensor. This is possible with an application of the perturbation method based on Taylor expansion and with the effective modules method. The computational procedure is implemented using plane strain analysis carried out with the finite element method (program MCCEFF) and the symbolic computations system MAPLE. The sensitivity gradients and probabilistic moments are commonly determined on the basis of partial derivatives for the homogenized elasticity tensor, calculated using the response function method with respect to some composite parameters. They are subjected separately to a normalization procedure (in deterministic analysis) and the relevant algebraic combinations (for the stochastic case). This enriched homogenization procedure is tested on a periodic fiber-reinforced two component composite, where the material parameters are taken as design variables and then, the input random quantities. The results of computational analysis are compared against the results of the central finite difference approach in the case of sensitivity gradients determination as well as the direct Monte-Carlo simulation approach. This numerical methodology may be further applied not only in the context of the homogenization method, but also to extend various discrete computational techniques, such as Boundary/Finite element and finite difference together with various meshless methods.     

不确定非线性结构动力响应的区间分析方法

研究多自由度非线性不确定参数系统的动力响应问题. 以区间数学为基础，将不确定 性参数用区间进行定量化，借助一阶Taylor级数，给出了近似估计非线性振动系统动力响 应范围的区间分析方法. 从数学证明和数值算例两方面，将其与概率摄动有限元法进行了比 较，结果显示区间分析方法对不确定参数先验信息具有要求较少、精度较高的优点.     

Change-of-variable interval stochastic perturbation method for hybrid uncertain structural-acoustic systems with random and interval variables

The response analysis from hybrid uncertain structural-acoustic systems with random and interval variables (HUSAS) plays an important role in the optimal design of structural-acoustic systems. In this work, a hybrid uncertain numerical method known as the change-of-variable interval stochastic perturbation method (CVISPM) is proposed to predict the interval of the response probability density function and the response confidence interval of a HUSAS. This method is based on perturbation analysis and the change-of-variable technique. In the proposed method, the response of a HUSAS is approximated as a linear function of random variables using the stochastic perturbation analysis. According to the approximated linear relationships between the response and the random variables, the change-of-variable technique is introduced to calculate the response probability density function. Based on the response probability density function, the interval perturbation approach is used to predict the interval of the response probability density function and the response confidence interval. A numerical example of a shell structural-acoustic system with random and interval variables was employed to verify the effectiveness and precision of the proposed method.     

Parameter perturbation method for dynamic responses of structures with uncertain-but-bounded parameters based on interval analysis

The study was intended to evaluate the range of dynamic responses of structures with uncertain-but-bounded parameters by using the parameter perturbation method. The uncertain parameters were modeled as an interval vector. The first-order perturbation quantities of responses of the perturbed system were obtained through the parameter perturbation method, and then taking advantage of interval mathematics a new algorithm to estimate the response interval was presented. Comparisons between the parameter perturbation method and the probabilistic approach from mathematical proofs and numerical simulations were performed. The numerical results are in agreement with the mathematical proofs. The response range given by the parameter perturbation method encloses that obtained by the probabilistic approach. The results also show good robustness of the proposed method.     

计算不确定结构系统静态响应的一种可靠方法

不确定性广泛存在于工程结构分析和设计过程之中，不能简单地予以忽略。目前，概率方法、模糊方法和区间方法是不确定性建模的三种主要方法。本文把具有不确定性的结构材料参数、几何参数和所受外力用区间数描述，通过求解线性区间方程组准确地计算了结构静态响应。计算结果易于扩张是区间计算的一个主要缺陷，本文提出了一种有效避免这一问题的方法。该方法把区间函数的计算和区间线性方程组的求解转化为相应的全局优化问题，来确定解中的每个区间元素的边界值，并采用一种智能性算法(实数编码遗传算法)来求解这些全局优化问题。本文首先采用数学和结构分析算例对该方法的正确性和有效性进行了验证，然后把该方法与有限元方法相结合计算不确定结构系统的响应范围，并和求解同类问题的方法进行了比较。     

Identification of stiffness, dissipation and input parameters of randomly excited non-linear systems: Capability of restricted potential models (RPM)

A dynamic identification technique in the time domain for time invariant systems under random external forces is presented. This technique is based on the use of the class of restricted potential models (RPM), which are characterized by a non-linear stiffness and a special form of damping, that is a product of the input power spectral density (PSD) matrix and the velocity gradient of a non-linear function of the total mechanical energy. By applying stochastic differential calculus and by specific analytical manipulations, some algebraic equations, depending on the response statistics and on the mechanic parameters that characterize RPM, are obtained. These equations can be used for the dynamic identification of the above mechanic parameters once the response statistics of the system to be identified are evaluated. The proposed technique allows one to identify single-degree-of-freedom or multi-degrees-of-freedom systems in the case of unmeasurable input. Further, the probabilistic characteristics of the external forces can be completely estimated in terms of PSD matrix.     

PROBABILISTIC DAMAGE IDENTIFICATION OF STRUCTURES WITH UNCERTAINTY BASED ON DYNAMIC RESPONSES

The probabilistic damage identification problem with uncertainty in the FE model parameters, external-excitations and measured acceleration responses is studied. The uncertainty in the system is concerned with normally distributed random variables with zero mean value and given covariance. Based on the theoretical model and the measured acceleration responses, the probabilistic structural models in undamaged and damaged states are obtained by two-stage model updating, and then the Probabilities of Damage Existence （PDE） of each element are calculated as the damage criterion. The influences of the location of sensors on the damage identification results are also discussed, where one of the optimal sensor placement techniques, the effective independence method, is used to choose the nodes for measurement. The damage identification results by different numbers of measured nodes and different damage criterions are compared in the numerical example.     

复合夹层梁瞬态响应的谱有限元法

采用谱有限元法进行复合夹层梁的瞬态响应分析．该方法基于复合夹层梁的六阶运动微分方程,以其波动解作为动力位移形函数,根据标准有限元策略来构建复合夹层梁的动刚度矩阵．在频域内,夹心粘弹性材料的频率相关性采用复模量模型来模拟,进而利用快速傅立叶变换技术(FFT),得到时域内复合夹层梁的瞬态响应分析结果．最后以两端固支夹层梁为例,对其进行了矩形脉冲荷载下的动力响应分析,并与通用有限元程序NASTRAN的计算结果进行了对比,两者吻合良好．