端部约束悬臂输流管道的非线性动力学特性

针对悬臂端受线性弹簧支承和扭转弹簧约束的约束悬臂输流管道,采用分岔图、相平面图、Poincare截面和Lyapunov指数等非线性振动的数值仿真方法,研究其在自激-参数激励-外激励联合激励作用下的非线性动力学特性,分析系统出现周期和混沌运动响应的参数条件,揭示其通向混沌的途径,探寻各参数对输流管道振动特性的影响规律和各参数之间的相互制约关系.数值仿真结果表明,管道系统随质量比、端部约束刚度和管道粘弹性系数的不同,分别呈现周期、概周期、阵发性和混沌运动多种响应形式,系统通过倍周期分岔或阵发性进入混沌,通过倍周期倒分岔脱离混沌.     

三维超材料的非线性电、磁响应特性

理论研究了埋入在非线性介质中的由金属线和谐振环构成的具有三维周期性结构的超材料的有效电、磁响应,计算了该种超材料的有效磁导率和介电常数实部和虚部随外场的变化关系.结果表明,这种超材料有效磁导率和介电常数的实部对外磁场的响应具有回滞效应.在一定条件下,这种结构的属性可在右手材料、电单负材料与磁单负材料之间转换,且这种转换可适当选择在某些特定频率范围内的外磁场强度的方式控制.     

一种基于支持向量机的非线性系统辨识方法

提出一种新的非线性系统辨识方法,基于支持向量机回归算法,选取高斯核函数构造了从输入空间到高维特征空间的非线性映射,以避免繁琐的运算,实现对非线性系统的辨识。仿真结果表明了SVM具有很好的拟合和泛化能力,同基于神经网络的非线性系统辨识相比,其辨识和泛化性能要优于神经网络。支持向量机的使用为工业过程的系统辨识提供了一条新的途径。     

吸收对准相位匹配线性电光效应的影响(英文)

研究吸收对准相位匹配线性电光效应的影响,得到适用于吸收介质的准相位匹配线性电光效应的耦合波方程及其解析解.计算结果表明,吸收不但削弱光的强度,且影响o光与e光之间的电光耦合.当o光与e光的吸收系数不同时,由于它们之间的电光耦合,总输出光强随外加电场变化而变化.研究结果可用于吸收介质的电光器件设计.     

梁削弱式节点连接的侧移钢框架柱计算长度系数

根据框架柱计算长度理论,通过引入梁削弱式节点参数特性,建立了四跨三层的梁削弱式节点连接的侧移钢框架数值模型.利用ANSYS软件对梁削弱式节点连接的侧移钢框架数值模型进行了非线性屈曲分析,得到了梁削弱式节点连接的侧移钢框架柱计算长度系数.结果显示,梁削弱式节点侧移钢框架柱计算长度系数较常规节点增加1.5%～12.9%.最后,经过回归得到了梁削弱式节点连接的侧移钢框架柱计算长度系数建议公式.     

Rapid prediction method for nonlinear expansion process of medical vascular stent

A neural network model with high nonlinear recognition capability was constructed to describe the relationship between the deformation impact factors and the deformation results of vascular stent.Then,using the weighted correction method with the attached momentum term,the network training algorithm was optimized by introducing learning factor η and momentum factor ψ,so the speed of the network training and the system robustness were enhanced.The network was trained by some practi-cal cases,and the statisti...     

Equilibrium equations for nonlinear buckling analysis of drill-strings in 3D curved well-bores

With the development of drilling technology, the oil/gas well has evolved from its early vertical straight form to the inclined, horizontal, plane curved, or even 3D curved well-bore. Understanding of the buck-ling behavior of a drill-string in a well-bore is crucial for the success of a drilling operation. Therefore, equilibrium equations for analyzing the buckling behavior of a drill-string in a 3D curved well-bore are required. Based on Love’s equilibrium equations for a curved and twisted rod in space, ...     

A macro-meso constitutive law for concrete having imperfect interface and nonlinear matrix

The overall behavior of concrete depends on its meso structures such as aggregate shape, interface status, and mortar matrix property. The two key meso structure characters of concrete, bond status of interface and nonlinear property of matrix, are considered in focus. The variational structure principle is adopted to establish the macro-meso constitutive law of concrete. Specially, a linear reference composite material is selected to make its effective behavior approach the nonlinear overall behavior of concrete. And the overall property of linear reference composite can be estimated by classical estimation method such as self-consistent estimates method and Mori-Tanaka method. This variational structure method involves an optimum problem ultimately. Finally, the macro-meso constitutive law of concrete is established by optimizing the shear modulus of matrix of the linear reference composite. By analyzing the constitutive relation of concrete established, we find that the brittleness of concrete stems from the imperfect interface and the shear dilation property of concrete comes from the micro holes contained in concrete. Supported by the National Natural Science Foundation of China (Grant Nos. 50679022, 90510017, 50539090) and National Basic Research Program of China (Grant No. 2007CB714104)     

随机环境下的幂变换门限ARMA-ARCH模型的极限行为

提出了随机环境下的幂变换门限自回归滑动平均自回归条件异方差模型,得出了其以几何速率收敛的充分条件．     

复合粒子群优化算法在模型参数估计中的应用

化工非线性模型的参数估计是较为困难的寻优问题，经典方法常会陷入局部极值。粒子群算法操作简便、容易实现且全局搜索功能较强，适用于非线性参数估计。但其参数值的确定与问题相关，若设定不当，会严重影响全局搜索的性能。今提出引入遗传算法，在粒子群算法的搜索过程中，逐代优选参数，包括惯性权值，加速常数，以此构建为复合粒子群优化算法。分析与测试表明，其全局搜索性能有显著改善。进一步的工作又将两种粒子群算法成功地应用于重油热解模型的参数估计。采用复合粒子群优化算法估计参数构建的重油热解模型，其预报相对误差比常规粒子群优化算法降低了8.97％，比简单遗传算法降低了23.21％，效果明显。