预测型数据挖掘中的优化方法

数据挖掘是一个多学科相互交叉融合所形成的具有广泛应用前景的新兴研究领域。本文从最优化的观点出发,论述了预测型数据挖掘中的优化问题,介绍了支持向量机优化模型与算法在模式分类和回归问题中的应用。最后指出从海量数据中进行数据挖掘为优化理论与方法带来的机遇和挑战。     

露天停车场停车位的优化设计

该文应用多种相关分析、综合评价和最优化模型等方法对露天停车场停车位进行优化设计。解决了停车场最优化规划问题以及在考虑停车场消防问题时的停车场最优化问题。该文综合最优化、非线性规划、多种相关分析等模型,结合MATLAB、 Excel等软件,对停车场优化问题进行了多角度的分析,并给出了合理的模型。在该文的最后对模型进行了评价,在实际应用中有较大的参考价值。     

基于多指标排序信息下Black-Litterman模型的研究

针对Black-Litterman模型中投资者观点难以量化的问题,本文提出一种基于TOPSIS方法的多指标排序信息下的Black-Litterman模型.首先通过TOPSIS方法对具有多个指标属性的资产进行排序,从而获得量化的观点.然后采用最新的随机最优化思想求解最优化问题,确定资产的组合权重.最后对构建的模型进行实证研究,选取上海证券交易所交易的十只个股进行资产配置.实证结果显示:与传统的投资组合方法相比,新方法可以有效的将价格以外的信息结合进来,从而提高了模型的稳健性和运用范围.     

动态规划在运动图像分析中的应用

对运动图像序列中的目标进行运动跟踪和估计,可以分别采用变形模型和弹性配准的方法,两问题的求解均可归纳为成本函数的最优化问题。提出了应用动态规划算法求解离散成本函数最优化的方法,把问题分解成多个子问题分别求解,并存储子问题的解以避免重复计算。该方法不仅可保证解的全局最优性,而且计算量小,有利于实现实时处理。在对临床X射线冠状动脉造影图像序列的实验中得到了满意的结果,匹配大约2000个点的血管骨架只需要一秒钟左右的时间。对运动场已知的模拟图像的实验证明运动估计误差小于1个像素（1像素=0．3mm）。     

一种流程型企业主生产计划优化方法研究

为了解决传统的流程型生产管理方法不适应现代市场需求的问题,提出了一种基于生产柔性提高的主生产计划优化方法.建立了流程型企业的产品需求与生产能力平衡的最优化模型,讨论了模型的求解方法--混合启发式算法.经研究表明,该方法在流程型生产管理理论和实践方面具有一定的参考意义.     

MEXCLP模型在优化公共设施选择问题上的研究

设施定位问题即UFL问题是NP-hard的组合优化问题,是聚类问题领域的热点问题之一,在数据挖掘和分类识别方面有着重要应用。针对设施定位问题,建立最大期望覆盖模型(MEXCLP),并给出求解该模型的启发式算法。     

空间数据立方体的物化视图选择方法研究

针对决策支持系统(DSS)中集成空间分析能力的应用趋势，研究基于空间数据仓库的一种决策分析工具——空间在线分析处理(Spatial OLAP)，拟解决影响空间OLAP在线响应的瓶颈难点——空间数据立方体的物化问题。首先系统地提出空间OLAP的模型，然后对现有空间度量物化视图选择方法进行改进，提出了双向空间Greedy算法。实验证明．该算法在降低选择时间和求解质量两方面具有更好的表现。     

基于灰色系统理论的时序数据挖掘技术

阐述了嵌入知识的数据挖掘思想和数据挖掘技术现状，结合灰色系统理论首次提出了时序数据挖掘的灰色系统方法集(GDMS)，并以灰色系统中的GM(1,1)模型为例，介绍了其具体算法。应用此算法对上海市2002～2005年的上网户数进行了预测。     

基于MGARCH模型的情景树生成算法(英文)

多阶段随机最优化模型的质量依赖于描述不确定环境的情景树的质量.本文从以下几个方面对现有的情景树生成算法进行了改进:为恰当反映随机数据过程高阶矩的变化,我们提出了一个基于MGARCH模型的新模拟方法来生成情景;为改进现有情景生成的序列最优化方法,我们用MGARCH模型来递归估计随机数据过程的中心矩,设计了一个新的混合智能算法来求解序列最优化方法中所遇到的非凸规划问题,并由此导出了一个基于MGARCH模型的、可用于生成一般结构多阶段情景树的新型有效序列最优化方法.最后,利用中国和美国股票市场的金融交易数据,我们进行了一系列数值试验以说明我们算法的实用性、灵活性和有效性.     

基于OPT的生产作业计划与控制系统设计

建立符合我国企业实际情况的生产作业计划与控制系统是我国企业现代化改造和适应新的生产方式的主要问题之一。论述了建立生产作业计划与控制系统模型的基本方法,以最优化生产技术（ＯＰＴ）的基本原理为基础,结合我国企业的实际情况,对现代制造系统生产作业计划与控制系统的编制方法进行了探讨和开发应用。     

Prediction of road accidents: A Bayesian hierarchical approach

In this paper a novel methodology for the prediction of the occurrence of road accidents is presented. The methodology utilizes a combination of three statistical methods: (1) gamma-updating of the occurrence rates of injury accidents and injured road users, (2) hierarchical multivariate Poisson-lognormal regression analysis taking into account correlations amongst multiple dependent model response variables and effects of discrete accident count data e.g. over-dispersion, and (3) Bayesian inference algorithms, which are applied by means of data mining techniques supported by Bayesian Probabilistic Networks in order to represent non-linearity between risk indicating and model response variables, as well as different types of uncertainties which might be present in the development of the specific models.     

基于Java3D的空间关联规则可视化原理与实现

空间数据挖掘可视化是空间数据挖掘研究的一个重要方面。可视化技术充分利用了图形和图像的表达能力以及人对于色彩和空间的敏锐的感知能力,使人机有机地融合在一起。本文在简述空间关联规则可视化基本需求的基础上,提出了空间关联规则可视化的一般方法,并进行用Java3D技术开发空间关联规则可视化工具的应用研究。重点在于分析利用Java3D技术进行空间关联规则可视化的基本原理和方法,并给出核心部分流程。最后将该技术应用于两个具体实例中,并给出实现的空间关联规则可视化界面。     

Parameter identification and shape optimization

Simulation of metal forming processes using the Finite Element Method (FEM) is a well established procedure, being nowadays possible to develop alternative approaches, such as inverse methodologies, in solving complex problems. In the present paper, two types of inverse approaches will be discussed, namely the parameter identification and the shape optimization problems. The aim of the former is to evaluate the input parameters for material constitutive models that would lead to the most accurate set of results respecting physical experiments. The second category involves determining the initial geometry of a given specimen leading to a desired final geometry after the forming process. The purpose of the present work is then to formulate these inverse problems as optimization problems, introducing a straightforward methodology of process optimization in engineering applications such as metal forming and structural analysis. To reach this goal, an integrated optimization approach, using a finite element code together with a numerical optimization program, was employed. A gradient-based optimization method, as a combination of the steepest-descent method and the Levenberg-Marquardt techniques, was used. Numerical applications in the parameter optimization category include, namely, the characterization of a non-linear elasto-plastic hardening model and the determination of the parameters for a nonlinear hyperelastic model. It is also discussed the simultaneous identification of both constitutive material model parameters and the friction coefficient parameters. From the point of view of shape optimization problems, the determination of the initial geometry of a specimen in a upsetting billing problem as well as a methodology for defining the most suited blank shape to be formed in a square cup, are discussed. The final results for both categories show that this kind of algorithms have great potential for future developments in more demanding and realistic benchmarks. It is also worth noting that the presented integrated methodology can be easily applied to a first introduction of optimization techniques and numerical simulation to undergraduate courses in engineering.     

悬立式深海钴结壳采矿头的参数化设计与优化

传统的螺旋滚筒式钴结壳采矿头在实际开采过程中会使较多的废石混入。为了能在复杂多变的海底矿区实现对钴结壳的精确剥离以及提高开采率,有必要设计一种可高效开采的采矿头。首先,针对海底钴结壳分布的地貌特征,提出了采矿头设计的基本要求并设计了一种悬立式采矿头,同时建立了其参数化设计的数学模型。然后,基于所设计的悬立式采矿头,以降低钴结壳破碎过程中的比能耗为目标,以采矿头截齿不发生干涉、合理布置截距、钴结壳产能达标及最大功率限制等为约束条件,建立了悬立式采矿头结构参数和工作参数的优化模型,并采用遗传算法进行求解。最后,为了验证优化模型的合理性,利用ABAQUS软件模拟了悬立式采矿头破碎钴结壳的过程,得到了优化前后采矿头破碎钴结壳的比能耗,并基于载荷波动特性对采矿头的稳定性进行了对比评价。理论计算结果表明,优化后悬立式采矿头破碎钴结壳的比能耗较优化前约降低了44%。仿真模拟结果表明,优化后悬立式采矿头破碎钴结壳的比能耗和所受载荷的波动系数较优化前分别降低了18%和31.3%,验证了优化模型的有效性。所设计的悬立式采矿头不仅在能耗、载荷波动方面有显著的改善效果,且能更好地适应海底钴结壳的分布特征,这可为深海钴结壳商业化开采的实现提供重要依据。     

An approach for the aging process optimization of Al–Zn–Mg–Cu series alloys

A new model based on least square support vector machines (LSSVM) and capable of forecasting mechanical and electrical properties of Al–Zn–Mg–Cu series alloys has been proposed for the first time. Data mining and artificial intelligence techniques of aluminum alloys are used to examine the forecasting capability of the model. In order to improve predictive accuracy and generalization ability of LSSVM model, a grid algorithm and cross-validation technique has been adopted to determine the optimal hyper-parameters of LSSVM automatically. The forecasting performance of the LSSVM model and the artificial neural network (ANN) has been compared with the experimental values. The result shows that the LSSVM model provides slightly better capability of generalized prediction compared to back propagation network (BPN) in combination with the gradient descent training algorithm. Considering its advantages of the computation speed, unique optimal solution, and generalization performance, the LSSVM model is therefore considered to be used as an alternative powerful modeling tool for the aging process optimization of aluminum alloys. Furthermore, a novel methodology hybridizing nondominated sorting-based multi-objective genetic algorithm (MOGA) and LSSVM has been proposed to make tradeoffs between the mechanical and electrical properties. A desirable nondominated solution set has been obtained and reported.     

PRELIMINARY OPTIMAL DESIGN OF CABLE-STAYED BRIDGES

A methodology for preliminary design of cable-stayed bridges using optimization techniques is presented. A simple model of the bridge, comprising beam elements for the longitudinal deck girder and lower and truss elements for stay cables, is used. The computer program SADDLE is used for optimization. Constraints are placed on stresses, displacements and member sizes under multiple loading conditions. The methodology enables the designer to vary the arrangement and size of stay cables and cross-sectional dimensions of the longitudinal deck girder. Consequently an aesthetically pleasing as well as an economical and feasible design of a cable-stayed bridge can be obtained quickly and efficiently.     

Elicitation and analysis of affective needs in vehicle design

This article describes a methodology for elicitation and analysis of affective needs for vehicle design. Driven by the concept of citarasa or emotional intent, the method has five steps. First, a model of emotional intent was conceptualised; second, a semantic framework of citarasa words was developed that mapped words to specific vehicle components to form a citarasa ontology; third, customer citarasa were elicited in the field using probe interview technique; fourth, affective needs were refined through Web survey; and fifth, the elicited citarasa were analysed using data mining techniques and the citarasa analysis tool. The tool is linked to the citarasa database that enables analysis of affective needs in several countries in Europe and Asia. The system has been technically verified, validated and tested for usability with consumers and automotive end-users.     

Decision-making and cooperative interaction via coupling agents in organizationally distributed systems

This paper develops a methodology for decision-making in organizationally distributed systems where decision authorities and information are dispersed in multiple organizations. Global performance is achieved through cooperative interaction and partial information sharing among organizations. The information shared among organizations is contrived using modified Lagrangian relaxation techniques. Novel to the methodology is that no single master problem with a global view of the system is required to guide the decision process. Rather, multiple artificial decision entities, termed Coupling Agents, are associated with subsets of coupling constraints. The proposed generic model can be applied to decision-making problems with a variety of mathematical structures. In this paper the methodology is applied to parameter design problems to illustrate the behavior of the proposed methodology in the realm of non-linear optimization.     

Optimization of machining techniques — A retrospective and literature review

In this paper an attempt is made to review the literature on optimizing machining parameters in turning processes. Various conventional techniques employed for machining optimization include geometric programming, geometric plus linear programming, goal programming, sequential unconstrained minimization technique, dynamic programming etc. The latest techniques for optimization include fuzzy logic, scatter search technique, genetic algorithm, Taguchi technique and response surface methodology.     

An integrated design and optimization environment for industrial large scaled systems

The objective of this paper is to demonstrate how a combination of design optimization theory and methodology can be applied to large-scaled industrial systems to efficiently improve their performance, reduce their costs or improve other design objectives. The scheme described was developed when other conventional design and optimization strategies failed in efficiently optimizing an air-to-air missile design for Lockheed Martin Missile and Fire Control. The efficient design scheme was developed for the system using a combination of optimization and design of experiment techniques. It will be shown that multidisciplinary design optimization techniques can be improved with a dependency-tracking demand-driven language resulting in an attractive choice for solving industrial type design problems. The design methodology holds true even for systems in which a large number of disciplinary design and analysis software are integrated. One reason for the efficiency of the scheme is the parameterized dependency-tracking environment in which the optimizations are carried out. With the hybrid approach developed, combining exploration and optimization techniques with the unique dependency-tracking and demand driven features of the environment, it was possible to reduce the computational time by as much as 44%. The design scheme developed and presented can be used to improve the design and optimization process for numerous other engineering applications.