混沌优化算法在不可分稳态大系统优化中的应用

讨论了整体目标函数关于各子系统不具有可加形式的大规模稳态系统的优化问题,将混沌优化算法应用于其最优值的求解,利用混沌运动的遍历性来得到优化问题的全局最优值.仿真结果表明,该算法简单易行,求解精度和可靠性较高,是解决不可分稳态大系统优化问题的一种有效方法.     

系统可靠度的优化模型及简单算法

提高系统可靠度，往往受到给定资源、重量、体积等的限制，如何在此限制下，进行单元可靠度的提高，使得系统的可靠度达到最大，这就是本研究的主要内容。     

约束下资本预算优化模型与算法

约束下资本预算是投资决策的常见问题,这类问题的通行解法不能确保获得真正的最优解,因此有必要运用运筹学方法来处理这类问题。     

灰色系统模型的优化岭回归算法

文献[1]指出了目前用普通最小二乘法估计灰微分方程参数的方法由于方程组的病态问题很难求解得合理的参数;文献[2]指出了根据初值求解灰色系统模型的时间响应式的方法由于初值的误差使所求得时间响应式产生系统误差.为了克服灰色模型的上述两个缺点,本文设计了一种求解灰色系统模型的优化岭回归算法,计算一个广泛引用的算例演示了这种算法的优越性.     

配气系统的统一优化模型及其参数规划法

本文为内燃机配气机构设计,建立了统一的优化模型,并探索出了有效的求解方法一参数规划法,能为不同型号,转速的配气机构提供最佳设计参数,便于生产部门采用.     

正态云模型果蝇优化算法及其应用

提出了一种基于正态云模型的果蝇优化算法(NCMFOA).该算法通过直接将果蝇位置赋值给气味浓度判定值和引入正态云模型来刻画果蝇嗅觉搜索行为的随机性与模糊性,从而解决了果蝇优化算法(FOA)不能搜索负值空间的缺陷,并有效克服了 FOA算法在解决复杂优化问题时容易陷入局部极值的不足.通过正态云模型熵值的动态调整,使得NCM...     

多周期多设备冰蓄冷系统的优化模型、算法及应用

本文以多周期多设备冰蓄冷系统运行的动态过程为主要约束,以设备运行状态等为离散优化变量,以设备流体流速等为连续优化变量,以系统运行总费用为目标函数,建立了非线性非光滑的混合整数规划,并论述了该规划问题最优解的存在性。依离散优化变量的有限性,把这个优化问题等价地分解成有限多个关于连续优化变量的线性规划,构造具体的优化算法。最后应用于一个实际冰蓄冷系统,表明了本文的数学模型及优化算法等的正确与有效性,达到缓解用电高峰的用电量,并降低了用户的运行费用。     

卫星舱内长方体群布局的优化模型及全局优化算法

本文研究了卫星舱内长方体群优化问题,建立了一个三维布局优化模型,并用图论,群论等工具克服了布局优化问题时断时续性质带来的困难,在此基础上构造了一个全局收敛的优化算法,文中所用的方法可用于求解类似问题。     

基于非线性规划的社会系统协调发展优化模型及其应用

社会系统协调发展优化问题是优化理论运用较少的领域.基于非线性规划方法,依据社会系统协调发展内涵,对社会系统协调发展目标函数做了具体改进,构建了社会系统协调发展优化模型,并根据协调发展类型分为超前型优化模型与滞后型优化模型.这两种优化模型的区别主要是约束条件的不同.依据优化模型的优化解,可以为某地区社会系统的协调发展提供清晰的调节与控制路径.最后利用上述方法对徐州地区物流基础系统、经济基础系统的协调发展进行了具体优化与调控.     

大系统试验选优理论研究

针对某些高维动态、大型线性、复杂非线性、复杂模拟仿真和定性知识模型，该文全面总结了作者采用数学模型、知识模型和试验选优理论相结合的方法而提出的试验选优理论，该理论体系可以使一些目前无法求解或很难求解的大型问题获得近似解。并在复杂水资源系统的优化中取得了良好的效果。     

超大规模集成电路布局的优化模型与算法

布局确定集成电路单元在芯片中的具体位置,在单元互不重叠的基础上优化一些性能指标。该问题是NP困难的组合优化问题,是超大规模集成电路物理设计的核心问题之一,对集成电路的性能指标,如线网可布通性、时延、功耗、电路可靠性等有重大影响。在现代的集成电路设计中,布局问题通常包含数百万个集成电路单元,以及大小相异的异质性模块,和各种复杂的布局约束。目前的超大规模集成电路布局算法通常分解为总体布局、布局合法化和详细布局三个步骤。根据近年来集成电路布局算法的研究进展,综述并分析集成电路的总体布局、布局合法化和详细布局的相关优化模型和算法,并展望进一步的研究方向。     

High-performance numerical algorithms and software for subspace-based linear multivariable system identification

Basic algorithmic and numerical issues involved in subspace-based linear multivariable discrete-time system identification are described. A new identification toolbox—SLIDENT—has been developed and incorporated in the freely available Subroutine Library in Control Theory (SLICOT). Reliability, efficiency, and ability to solve industrial identification problems received a special consideration. Two algorithmic subspace-based approaches (MOESP and N4SID) and their combination, and both standard and fast techniques for data compression are provided. Structure exploiting algorithms and dedicated linear algebra tools enhance the computational efficiency and reliability. Extensive comparisons with the available computational tools based on subspace techniques show the better efficiency of the SLIDENT toolbox, at comparable numerical accuracy, and its capabilities to solve identification problems with many thousands of samples and hundreds of parameters.     

Volterra-type models for nonlinear systems identification

In this work, multi-input multi-output (MIMO) nonlinear process identification is dealt with. In particular, two Volterra-type models are discussed in the context of system identification. These models are: Memory Polynomial (MP) and Modified Generalized Memory Polynomial (MGMP), which can be considered as a generalization of Hammerstein and Wiener models, respectively. Both of them are appealing representations as they allow to describe larger model sets with less parametric complexity. Simulation example is given to illustrate the quality of the obtained models.     

Optimization of finite element bidimensional models: an approach based on genetic algorithms

This paper deals with the optimization of 2D finite element shapes using the very promising methods based on genetic algorithms. The codification of the design variables is carried out by generating series of strings in binary code. Classical genetic operators such as crossover, mutation and reproduction are used for the optimization process. A more refined operators needed to improve the performance of the process are used as well. Some illustrative examples are presented and discussed     

大规模系统的全局优化

本文讨论了可分非凸大规模系统的全局优化控制问题 .提出了一种 3级递阶优化算法 .该算法首先把原问题转化为可分的多目标优化问题 ,然后凸化非劣前沿 ,再从非劣解集中挑出原问题的全局最优解 .建立了算法的理论基础 ,证明了算法的收敛性 .仿真结果表明算法是有效的 .     

On consistency of recursive least squares identification algorithms for controlled auto-regression models

The recursive least squares (RLS) algorithms is a popular parameter estimation one. Its consistency has received much attention in the identification literature. This paper analyzes convergence of the RLS algorithms for controlled auto-regression models (CAR models), and gives the convergence theorems of the parameter estimation by the RLS algorithms, and derives the conditions that the parameter estimates consistently converge to the true parameters under noise time-varying variance and unbounded condition number. This relaxes the assumptions that the noise variance is constant and that high-order moments are existent. Finally, the proposed algorithms are tested with two example systems, including an experimental water-level system.     

可列非齐次隐Markov模型的强大数定律

隐马尔科夫模型被广泛的应用于弱相依随机变量的建模,是研究神经生理学、发音过程和生物遗传等问题的有力工具。研究了可列非齐次隐 Markov 模型的若干性质,得到了这类模型的强大数定律,推广了有限非齐次马氏链的一类强大数定律。     

Parametric solution to the joint system identification and optimization problem

This study is concerned with the simultaneous identification and optimization of static systems. The necessity and the advantages of an integrated approach to the identification and optimization of the system model is established theoretically as well as computationally. A parametric approach to the integrated problem is proven to converge to the integrated problem solution. The general methodology of decomposition of large-scale systems is extended by implementingfeasible decomposition of the joint problem. A multilevel approach is then utilized to successfully solve example problems. Handling the system constraints via a penalty-function technique is shown to be an efficient approach when using the parametric formulation of the joint problem. Numerical results for two example problems are presented using the Univac 1108 digital computer, revealing the economic advantages and disadvantages of the integrated approach to the identification and optimization problems.The authors wish to thank Messieurs I. Lefkowitz, L. S. Lasdon, F. Gembicki, and O. B. Olagundoye for their critique, comments, and suggestions during the course of this study.     

Decomposition in large system optimization using the method of multipliers

Although the method of multipliers can resolve the dual gaps which will often appear between the true optimum point and the saddle point of the Lagrangian in large system optimization using the Lagrangian approach, it is impossible to decompose the generalized Lagrangian in a straightforward manner because of its quadratic character. A technique using the linear approximation of the perturbed generalized Lagrangian has recently been proposed by Stephanopoulos and Westerberg for the decomposition. In this paper, another attractive decomposition technique which transforms the nonseparable crossproduct terms into the minimum of sum of separable terms is proposed. The computational efforts required for large system optimization can be much reduced by adopting the latter technique in place of the former, as illustrated by application of these two techniques to an optimization problem of a chemical reactor system.The authors would like to acknowledge the valuable comments given by Professor D. G. Luenberger of Stanford University.