基于遗传算法和Matlab的一种可靠度计算方法

利用由Matlab语言编制的遗传算法优化工具箱计算工程结构的可靠度，给出了设计验算点，避免了非线性功能函数的泰勒级数展开及遗传算法程序的编制。     

基于无网格NS-PIM与有限元法的边坡应变能及位移上下限解研究

为尽可能精确求解边坡应变能及位移的上下限解的问题,结合无网格NS-PIM方法与有限元法,求解出了离散精度不同的边坡模型的应变能及位移,快速确定了精确解的分布范围,并依据工程精度要求,进一步细化模型以使计算结果逼近精确解。实例计算表明,NS-PIM方法、有限元方法可分别得到应变能和总体位移的上、下限解,为位移监测分析和边坡稳定性评价提供了良好思路。     

湍流_转捩模型对气冷涡轮气热耦合计算影响的对比研究

开发了基于三维非结构网格的导热求解程序,将数值解与解析解对比验证了加权最小二乘法相对于格林函数法在求解梯度时具有更高的精度,将基于加权最小二乘法的三阶精度的导热程序与叶栅流场计算程序HIT-3D耦合,实现了气热耦合计算,对MARKII叶片5411实验工况进行了数值模拟,并借助于HIT-3D程序具备的湍流模型研究了转捩对传热计算的影响。结果表明:BL模型预测的温度与实验值相差最大,在转捩点相差10%,BL+AGS转捩模型、SST-Gama模型、q-ω模型由于具备模拟转捩的能力,因此在转捩区计算的温度误差较小,约5%,可见考虑转捩的湍流模型能更好地预测涡轮叶片的温度。     

考虑非线性强度指标的面板堆石坝坝坡稳定可靠度分析

采用非线性强度指标按简化毕肖普法给出的抗力弯矩和滑动弯矩定义坝坡稳定可靠度分析的功能函数,并应用一次二阶矩法计算可靠指标,提出了一种混凝土面板堆石坝坝坡稳定可靠度计算方法,以某面板坝坝坡为例进行了稳定可靠度分析。结果表明,各工况上、下游坝坡及深层滑动的最小安全系数均满足《碾压式土石坝设计规范》的要求,最小可靠指标除地震工况下游坝坡外都满足《水利水电工程结构可靠度设计统一标准》的要求;安全系数和可靠度计算结果的总体规律一致,但在某些情况下有较大差异。对于重要工程,建议同时进行安全系数和可靠度分析。     

Green-Ampt入渗模型的一种显式近似解

为解决Green-Ampt(GA)入渗模型只有隐格式解的问题,给出了高精度的显格式近似解,该显格式构造一组幂函数作为基函数,通过最小化近似解与精确解之间的误差,达到对GA模型的逼近。在对误差函数最小化的过程中,首先证明优化解的存在性,同时采用自适应的优化分段方法,在满足精度的条件下减少基函数的级数,达到简化计算的目的。针对入渗时间趋于无穷导致近似解发散问题,建立了修正公式,确保了该逼近方法的全局收敛性。通过与其他近似解的比较,验证了该方法的稳定性和精确性。     

基于改进模糊综合评价法的太湖富营养化评价

针对传统模糊综合评价法构造隶属函数和计算权重方法存在的一些问题,提出将隶属函数改为非线性的S函数,并将指标值的离散程度添加到计算权重中,从而使模糊综合评价法更适用于富营养化评价。分别采用改进的模糊综合评价法、灰色关联度法、Hopfield神经网络模型对太湖水质进行富营养化评价,结果表明这三种方法评价结果接近,但当评价值接近于某级别标准值时,可能会相差一个等级,因此必要时可采用多种评价方法进行全面研究。     

基于散射参数的电压互感器高频数学模型

提出了一种基于电压互感器（PT）散射参数的VFTO下PT数学模型的建立方法。用Agilent4395A网络分析仪测量PT的散射参数,利用测量得到的散射参数计算PT的电压传递函数,然后利用矢量匹配法对计算得到的电压传输函数进行有理函数逼近得到电压传递函数,再利用测量得到的PT的一次电压和传递函数进行递归卷积计算可以得到仿真的二次电压,把测量的二次电压和仿真得到的二次电压进行对比,验证了该数学模型的准确性。     

消除混凝土坝应力网格效应的三次等效应力法

针对混凝土坝有限元等效应力在角缘部位的失真问题,基于挡水结构的弹性力学解,提出一种改进三次等效应力法,即选取三次多项式模拟混凝土坝水平截面的正应力分布,确定多项式的待定系数,得到三次等效应力分布形式。以混凝土重力坝和拱坝为例,分别以线弹性有限元法、线性等效应力法、改进的三次等效应力法计算混凝土坝建基面的法向正应力和主应力,结果表明改进的三次等效应力法计算结果不仅有效减小了对计算网格的依赖性,且与线弹性有限元计算结果相当接近。结果可为混凝土坝的安全性评价提供参考。     

基于扩展二次锥规划的最优潮流模型研究

针对最优潮流(Optimal Power Flow,OPF)随着系统规模的增加出现求解效率偏低问题,建立了一种基于扩展二次锥规划(Extended Second-Order Cone Programming,ESOCP)的最优潮流矢量化模型。该模型由非线性OPF变形得到。其中节点功率方程等价转换为线性表达式,在采用原始-对偶内点法对其求解时,可以有效降低海森阵的稠密度,提高求解效率,并能保持内点法优良的求解特性。同时利用电力网络的支路节点关联关系对模型约束条件进行预处理,减少约束和变量数,令算法计算性能和求解效率得到大幅提高。通过对IEEE-4到300节点5个标准系统以及一个1047实际系统进行仿真计算表明:ESOCP-OPF模型与变形前OPF模型计算结果一致,说明所提模型是正确且有效的。所提预处理方法进一步提高计算效率,且对大规模系统更有效,具有较好的实际应用价值和前景。     

复杂地基下重力坝抗滑稳定可靠度分析

针对复杂地基条件下重力坝不同抗滑稳定评估方法的适应性问题及传统刚体极限平衡法中多滑面功能函数难以显式建立的现状,在传统安全系数评估方法的基础上,构建了失效通道的抗滑稳定功能函数,结合响应面方程和验算点法进行抗滑稳定可靠度分析,提出了考虑失效模式间相关性的基于条件概率降维法（DRMCP法）的复杂地基重力坝多通道失效的串联体系可靠度计算方法,实现了对复杂地基情况下的高混凝土重力坝抗滑稳定安全可靠性进行分析。某高重力坝实例分析结果表明,地基材料的复杂性、变异性造成同一滑移通道安全系数与可靠度并非完全对等,复杂地基情况下体系可靠度主要取决于最薄弱通道的单点可靠度。     

预应力砼受弯构件裂缝控制的理论分析

总结了预应力混凝土梁裂缝控制的几种方法,提出了用简单名义拉应力表在宾PPC结构构件裂缝控制及验算建议,分析了我国目前规范对抗裂要求的具体条款并提出改进建议。     

Efficient and reliable analysis of structural collapse by using a DQFEM along with a GSR-based accelerated equilibrium iteration procedure

The differential quadrature finite element method which uses the differential quadrature techniques to the finite element discretization is used to analyze the static linear or non-linear elastic–plastic collapse of structural problems. The incremental loading procedure in conjunction with the use of equilibrium iteration is used to update the response history. The equilibrium iteration can be carried out by using the global secant relaxation-based accelerated constant stiffness iteration procedure which is efficient and reliable even the load stage is up to the near collapse level. Sample problems are analyzed. Numerical results demonstrate the algorithm.     

坝后背管弹性中心法与有限元方法对比分析

结合某水电站坝后背管，采用结构力学弹性中心法和有限元法分析了内水压力作用下钢衬和环向钢筋的应力及管道径向裂缝，对其计算结果进行对比分析，说明两方法的实用性和有效性。     

Analysis of Nonlinear Characteristics for Thermoelastic Half Plane with Voids

Based on the linear theory of thermoelastic materials with voids, a nonlinear mathematical model for thermoelastic half-plane problems with voids is presented under the finite deformation. The differential quadrature method and finite difference method are applied to discrete governing equations on the spatial and temporal domain, respectively, and a set of nonlinear discretization equations is yielded and solved by Newton-Raphson method. The validity and convergence of the presented method are examined by using comparative analysis. The characteristics for half-plane problems with three kinds of thermoelastic materials with voids are analyzed to study the effect of voids and temperature.     

Thermal Postbuckling Analysis of Functionally Graded Beams

Thermal buckling and postbuckling analysis of functionally graded (FG) beams is presented. The governing equations are based on the first-order shear deformation beam theory (FSDT) and the geometrical nonlinearity is modeled using Green's strain tensor in conjunction with the von Karman assumptions. For discretizing the governing equations and the related boundary conditions differential quadrature method (DQM) as a simple and computationally efficient numerical tool is used. Based on displacement control method, a direct iterative method is employed to obtain thermal postbuckling behavior of FG beams with different boundary conditions and geometrical parameters.     

流动数值模拟中GDQ方法的应用

为了满足微电子机械系统的飞速发展,探讨了一种程序简单、计算成本较低的微尺度流动数值模拟新方法-广义微分积分方法.研究工作根据微尺度库埃特流动特点对N-S方程加以改进,建立起相应的理论模型,采用这个模型流动模拟能拓宽到滑移流动区和过渡流动区.然后采用广义微分积分方法来进行流场数值模拟,并用蒙特卡洛直接模拟方法(DSMC)加以验证,结果表明利用新建理论模型的模拟结果与验证结果基本吻合,但是也发现随着努森数的增加两者的误差增加,表明新建理论模型还有必要进一步完善.     

Thermal Buckling Analysis of a Mindlin Rectangular FGM Microplate Based on the Strain Gradient Theory

This article is aimed at developing a nonclassical Mindlin rectangular functionally graded material (FGM) microplate based on the strain gradient theory (SGT) to study the thermal buckling behavior of microplates with different boundary conditions. This theory comprises material length scale parameters to interpret size effects. The developed model encompasses classical and modified couple stress Mindlin microplate models, if all the material length scale parameters or two of them are taken to be zero, respectively. The Mindlin rectangular FGM microplate is considered to be made of a mixture of metal and ceramic of which the volume fraction is described through a power low function. According to Hamilton's principle and the generalized differential quadrature (GDQ) method, the stability equations and associated boundary conditions are obtained and discretized, respectively. Current formulations provide a possibility to have all types of boundary conditions which herein, FGM microplates with three commonly used boundary conditions are considered. Three different types of thermal loads including uniform, linear and nonlinear temperature rises along the thickness of FGM microplates are considered. The dimensionless critical buckling temperature difference (DCBTD) predicted by SGT is compared with that of modified couple stress theory (CST) and classical theory (CT) which it is found that CST and CT underestimate the DCBTD. Also, effects of the boundary conditions, length scale parameter and material gradient index of FGM microplates on the DCBTD are judiciously investigated.     

乏气送粉锅炉煤粉浓度软测量技术及其仿真研究

锅炉燃烧过程中,喷燃器出口煤粉浓度不均将会导致炉膛火焰中心偏斜,从而引起炉膛气流冲刷后墙及右墙,高温过热器,高温再热器出现局部超温,结焦的现象。因此准确测量各风管中的煤粉浓度并指导调节对锅炉的安全,经济运行非常重要,为此,必须寻找一种简单,高效,实用而且适合工程应用的煤粉浓度测量的方法。对于热风送粉锅炉的煤粉浓度测量,国内外做了大量的研究,并得到了工程应用。本文提出了一种乏气送粉方式下基于气固两相流理论,根据风粉混合前后压力差大小计算粉煤浓度的新方法,并在理论推导的基础上进行了仿真研究,结果表明,煤粉浓度计算值与混合压差呈很好的对应, 说明能量法理论计算公式是适用的。     

On the discrete ordinates method for radiative heat transfer in anisotropically scattering media

In the discrete ordinates method (DOM), the normalized condition for the numerical quadrature of some complex scattering phase functions may not be satisfied. In this paper, a revised discrete ordinates method (RDOM) is developed to overcome this problem, in which a renormalizing factor is added into the numerical quadrature of in-scattering term. The RDOM is used to solve the radiative transfer problem in one-dimensional anisotropically scattering media with complex phase function. The radiative heat fluxes obtained by the RDOM are compared with those obtained by the conventional discrete ordinates method (CDOM) and Monte Carlo method. The results show the RDOM can overcome the false scattering resulted from the numerical quadrature of in-scattering term and improve largely the accuracy of solution of the radiative transfer equation by comparison with the CDOM.     

A numerical investigation on the heat and fluid flow of various nanofluids on a stretching sheet

Following the necessity of investigating fluid flow and heat transfer in the stretching sheet problem and effect of nanofluids on them, performance of various nanofluids were investigated in the present study. Three base fluids (deionized water, ethylene glycol, and engine oil) in combination with 18 nanoparticles (metals and their oxides) were investigated. While experimental methods are preferable, a mathematical model was developed and solved by applying differential quadrature method due to lack of such experimental data. With the results obtained in the real dimensions, the error caused by the cancellation of the viscosity effect due to the dimensionless variables was omitted. Effects of magnetic field and volume fraction of nanoparticle on the fluid flow and heat transfer characteristics were investigated. Highest heat transfer rate as well as small amounts of shear stress was obtained for deionized water–Al and deionized water–Mg nanofluids. Increasing volume fraction of nanoparticle was observed to increase both heat transfer and shear stress rates, while presence of a magnetic field caused an increase in shear stress and decrease in heat transfer rate.