非线性优化

The main goal of this dissertation is to develop and analyze interior-point algorithms and their variants for solving semi-definite, quadratic, combinatorial and nonlinear optimization problem that arise in production management, resource allocation, communication networks and computer science.

本文主要讨论和分析了内点算法的基本理论和基本思想，并将其推广来求解一般的半正定规划、二次规划、组合优化以及非线性规划问题，这些问题在工业生产管理、资源分配、网络通讯和计算机技术等领域中是会经常碰到的。

Keywords:steel structure roof; steel tube truss; large cantilever; structure analysis; geometric nonlinear; optimization design

关键词］钢结构屋盖；钢管桁架；大悬挑；结构分析；几何非线性；优化设计

In the other hands, we present a levelvalue estimation method for solving global optimization. The global optimization problem will be converted into solve the first root of a nonlinear equation.

此外，我们提出了求解总极值问题的水平值估计方法，将一全局优化问题转化为求解一元非线性函数方程的根。

The root-mean-square roughness of STO thin films with thickness of 300 nm was only 0.25 nm.

这些研究结果为优化STO薄膜非线性介电性质提供了必要的理论依据。

Based on the similarity principle and nonlinear finite element method, reposition balance method was put forward to find the prestress.

基于相似原理结合非线性有限元法提出了膜结构找力分析的复位平衡法，根据形面的拓扑关系对复位平衡法进行了优化。

In chapter 3, prediction theory of time series is reviewed, then non-linear autoregressive model of time series-feedforward neural networks is presented, at the same time a new method-SVD is proposed to design the architecture of feedforward neural networks. By using a trained network with a sufficient number of hidden nodes, according to the distribution of singular values of hidden nodes outputs, we can decide the optimum number of hidden nodes, then the reduced network will be retrained.

第三章：先回顾了时间序列的预测理论，提出了时间序列的非线性自回归模型-前馈神经网络模型，并提出了一种设计前馈神经网络的新方法-奇异值分解法，初始网络结构足够大，网络训练后，对网络隐层节点输出所形成的矩阵作奇异值分解，根据奇异值分布情况，决定网络隐层节点的个数，优化后的网络重新开始训练。

To improve the robust characteristic s , the paper makes use of nonlinear constraint method is used to optimize the weight vector of adaptive pattern. The final solution of weight vector is different from alterable diagonal loading and the unknown parameter in optimum optimal solution can be obtained easily and accurately.

为了提高算法的稳健性，本文采用非线性约束方法对进行优化，的优化解在形式上不同于可变对角加载类算法，且优化解中的待定参数容易准确求出。

Aimed at the typical defects of sample resource lacking and easy local optimal convergence, in this paper, a new way of inserting ANN into genetic algorithm is implemented, which make full use of the nonlinear solution calculation ability of ANN and the full field optimal solution seeking ability of GA. The method optimized the tube stagger spinning parameters.

针对神经网络中样本资源紧张，易于收敛于局部优解的问题，本文将神经网络模型嵌入遗传算法中，充分利用了神经网络的非线性求解能力和遗传算法的全局寻优能力，实现了对错距旋压工艺参数的智能优化。

Genetic algorithm is stochastic searching and optimization method using concept of natural evolution and genetics.

遗传算法是一种模仿生物自然进化过程的随机搜索和优化算法，其优势在于可以高效的处理传统搜索方法难以解决的非线性问题。

And the structure reliability formula based on the improved vertex method was suit to multi-fuzzy variable and non-linear system.

针对模糊－模糊干涉模型和线性结构系统应用一般的顶点法给出了求解结构模糊可靠度计算公式；对具有多模糊变量非线性结构系统用改进的顶点法优化模型给出了求解结构模糊可靠度计算公式。

It has been put forward that there exists single Ball point and double Ball points on the symmetrical connecting-rod curves of equilateral mechanisms.

从鲍尔点的形成原理出发，分析对称连杆曲线上鲍尔点的产生条件，提出等边机构的对称连杆曲线上有单鲍尔点和双鲍尔点。

The factory affiliated to the Group primarily manufactures multiple-purpose pincers, baking kits, knives, scissors, kitchenware, gardening tools and beauty care kits as well as other hardware tools, the annual production value of which reaches US$ 30 million dollars.

集团所属工厂主要生产多用钳、烤具、刀具、剪刀、厨具、花园工具、美容套等五金产品，年生产总值3000万美元，产品价廉物美、选料上乘、质量保证，深受国内外客户的青睐