混合递阶遗传算法优化小波神经网络模型的研究

本文提出了一种混合递阶遗传算法用于优化小波神经网络模型,并通过对这种神经网络模型的优化仿真,证明了该法在解决多输入多输出的非线性逼近及缓解多输入多输出的"维数灾"问题上的有效性.     

Multi-criteria genetic optimization for distribution network problems

This paper develops a multi-criterion genetic optimization for solving distribution network problems in supply chain management. Distribution problems deal with distribution from a number of sources to a number of destinations, in which various decision factors are closely related and influence each other. Genetic algorithms have been widely adopted as the optimization tool in solving these problems. This paper combines analytic hierarchy processes with genetic algorithms to capture the capability of multi-criterion decision-making. The proposed algorithm allows decision-makers to give weightings for criteria using a pairwise comparison approach. The numerical results obtained from the new approach are compared with the results obtained from linear programming. The result shows that the proposed algorithm is reliable and robust. In addition, it provides more control for decision-makers on the determination of the optimization solutions, and gains more information for a better insight into the distribution network. This revised version was published online in October 2004 with a correction to the issue number.     

计及电动汽车充电站的配电网综合优化研究

针对电动汽车充电站接入配电网优化运行这一问题,为达到降低配电网网损和提升网络节点电压质量的目标,在常规的配电网优化措施基础上,就计及电动汽车充电站接入的配电网综合优化问题进行了研究.分析了网络重构、无功优化、电动汽车充电站注入功率三者之间的关系,提出了应用粒子群算法（PSO）对三者进行统一优化的方法,并在IEEE33配电系统上进行了仿真试验.研究结果表明,在综合优化条件下系统的网损最低、电压质量最佳,从而验证了该方法的有效性.     

An artificial neural network approach to multiple-response optimization

In many manufacturing cases, engineers are required to optimize a number of responses simultaneously. A common approach for the optimization of multiple-response problems begins with using polynomial regression models to estimate the relationships between responses and control factors. Then, a technique for combining different response functions into a single scalar, such as a desirability function, is employed and, finally, an optimization method is used to find the best settings for the control factors. However, in certain cases, relationships between responses and control factors are far too complex to be efficiently estimated by polynomial regression models. In addition, in many manufacturing cases, engineers encounter qualitative responses, which cannot be easily stated in the form of numbers. An alternative approach proposed in this paper is to use an artificial neural network (ANN) to estimate the quantitative and qualitative response functions. In the optimization phase, a genetic algorithm (GA) is considered in conjunction with an unconstrained desirability function to determine the optimal settings for the control factors. Two manufacturing examples in which engineers were asked to optimize multiple responses from the semiconductor and textile industries are included in this article. The results indicate the strength of the proposed approach in the optimization of multiple-response problems.     

使用PSO寻优的遗传算法在配网重构中的应用

配电网具有闭环设计、开环运行的特点.提出了一种新的使用PSO加速寻优的遗传算法用于配电网重构,以减少网损.在重构优化过程中,采用了高频变异和消亡算子,能有效维持种群的多样性,避免算法早熟收敛.同时利用PSO更新个体的速度和位置,提高了收敛速度.对69节点系统重构,研究结果表明提出的算法具有较高的计算效率.     

一种交通信息采集传感器网络任务分配方法

针对城市道路交通信息采集传感器网络面向复杂交通参数协同采集的任务分配问题,将传感器网络映射为多Agent系统,以任务完成时间、节点能耗和网络负载平衡度作为评价函数,采用基于联盟的协同方法,构造传感器网络任务分配的非线性多目标优化模型.采用遗传模拟退火算法搜索最优联盟结构,实现任务分配策略优化.在道路交通信息采集实际场景中进行仿真实验,结果表明,遗传模拟退火算法能够有效地优化任务分配的联盟结构,与其他优化算法相比,优化的模型适应度函数值低,任务完成时间短,网络能耗小.该方法能够用于面向交通信息采集传感器网络的协同检测任务分配问题.     

开放式动态网络车辆路径问题的粒子群算法

针对现有车辆路径问题模型对动态性和开放性的约束限制,建立了开放式动态网络车辆路径的数学模型,使用连续时间依赖函数表示动态网络,并提出了基于惯性权重自适应调整和状态分类更新的粒子群算法求解该问题。根据社会认知理论,每个粒子依据当前位置与种群最优位置和自身历史最优位置的相对关系,动态调整自身的惯性权重。为避免早熟收敛,增加群体的多样性,使用分类更新策略。对于优秀的粒子,通过计算信息熵,使用特殊的状态更新公式计算其状态;对于适应度低的粒子,通过公告板统计出现的频率,进行粒子更新。通过实验仿真,对算法的参数进行了分析,并通过与其他算法的比较,验证了该算法的有效性。     

基于复杂网络理论的多项目实施资源优化配置研究

针对多项目实施过程中的资源优化配置问题,首先建立了基于复杂网络理论的多项目实施资源配置复杂网络模型,并结合Pajek分析软件对模型参数进行分析研究;其次提出了资源受限下的多项目实施资源配置算法;最后以多项目实施中资源配置为例,结合资源优化配置算法,以资源配置效益最大化为目标函数,对资源受限下的多项目实施资源优化配置问题进行研究,验证了该方法的可行性和有效性.     

碳税政策下分销网络选址—路径问题的鲁棒优化

为了促进节能减排,提出碳税政策下分销网络设计中的多容量等级选址—路径鲁棒优化问题.通过引入bo x不确定集合与两类不确定水平参数来描述单位运费的不确定性.采用混合整数非线性规划对问题进行数学描述,进而基于强对偶理论将非线性模型转化为线性化的鲁棒等价模型.采用GUROBI软件进行了数值计算和分析.数值分析表明:碳税政策具...     

基于GPRS的配电网动态无功优化控制策略研究

配电网无功优化控制是解决配电网电能损耗大、电压水平低这一问题的有效手段,采用动态无功优化求解时一般按照各节点全天短期预测负荷曲线合理分段优化,即将动态无功优化控制转化为多时段静态无功优化问题.针对现有配电网实际负荷的不规则变化会给该优化方法带来无功过补偿和欠补偿的问题,提出了一种基于实测负荷数据和预测负荷数据的动态无功优化混合控制策略;为了提高配电网无功优化的计算速度,使优化算法更好地适用于实际在线环境,引入了序优化理论框架,采用了改进的禁忌搜索算法作为无功优化算法;通过对34节点配电系统进行了算例分析.研究结果表明,该控制策略和优化算法在实用性、可行性和计算速度等方面均有良好表现.     

Topology optimization for lightweight cellular material and structure simultaneously by combining SIMP with BESO

This paper presents a hybrid algorithm for topology optimization of lightweight cellular materials and structures simultaneously by combining solid isotropic material with penalization (SIMP) and bi-directional evolutionary structural optimization (BESO). Microstructure of the lightweight cellular material is assumed unique in the structure to make the proposed method feasible. A new sensitivity analysis formula with respect to the discrete variable is derived by a principal submatrix stiffness matrix, by which the material can be effectively removed from or added to cellular. Moreover, the validity of the proposed method is then demonstrated through two numerical examples (a simple supported beam and a cantilever beam), which can be easily applied in a variety of practical situations.

     

结合深度卷积网络与加速鲁棒特征配准的图像精准定位

针对在大图像中定位小块区域图像的需求,本文提出一种结合深度卷积网络与加速鲁棒特征(SURF)配准的精准定位方法。将标准大区域图像分割成若干个小参考图像,利用深度卷积网络和类局部敏感哈希降维法提取参考图像集的特征并形成特征库;基于特征库,提出了先检索多个相似参考图像后再进行SURF精确配准的两阶段方法,实现目标小图像在标准大图像中的定位。针对电子工业过程中高密度柔性电路板(FPC)及精确末制导中的图像定位数据进行实验,实验结果表明,该方法避免了传统SURF算法大量的特征提取与配对过程,SURF特征提取数减少近90%;与直接根据图像特征进行配准的传统定位方法相比,在保证定位准确率的基础上,耗时可缩小一个数量级以上。     

采用正交函数优化三维自由曲面

自由曲面重建时,一般采用增加控制点个数来提高曲面拟合精度,但控制点的增加会引起曲面连续阶数变低.通过分析传统算法在曲面拟合中的不足,提出采用正交函数的曲面拟合方法,该算法避开不规则数据点进行拟合,保证了曲面的光顺性.当采用3×3阶曲面,36个控制点进行曲面拟合时,曲面重构最大误差仅为0.077 0mm,曲面满足G2曲率连续.实验结果表明,所提出的曲面优化算法,实现了曲面控制点数与曲面光顺和重构精度的最佳匹配,提高了曲面拟合的品质.     

Study on the tolerance allocation optimization by fuzzy-set weight-center evaluation method

The tolerance allocation optimization method by fuzzy-set weight-center evaluation is used to derive the manufacturing difficulty coefficient, through quantifying the manufacturing condition factors which all affect the manufacturing cost, including the forming means of the blank, element size, machining surface features, operator’s skills, and material’s machinability. The coefficient is then converted into a weight factor used in the inversed square model representing the relationship between the cost and tolerance, a cost-objective function model based on optimal tolerance allocation according to the manufacturing conditions is thus established for optimizing and allocating the tolerances. Integrating this model into computer-aided-tolerance-allocation makes it more convenient, accurate, and feasible.     

A neural network approach for prediction of discharge in straight compound open channel flow

Most natural rivers and streams consist of two stage channels known as main channel and flood plains. Accurate prediction of discharge in compound open channels is extremely important from river engineering point of view. It helps the practitioners to provide essential information regarding flood mitigation, construction of hydraulic structures and prediction of sediment load so as to plan for effective preventive measures. Discharge determination models such as the single channel method (SCM), the divided channel method (DCM), the coherence method (COHM) and the exchange discharge method (EDM) are widely used; however, they are insufficient to predict discharge accurately. Therefore, an attempt has been made in this work to predict the total discharge in compound channels with an artificial neural network (ANN) and compare with the above models. The mean absolute percentage error with artificial neural networks is found to be consistently low as compared to other models.     

An artificial neural network approach on parametric optimization of laser micro-machining of die-steel

In the present research, laser micro machining (LMM) of tungsten-molybdenum general purpose high speed steel (Rex M2) has been studied. Selection of optimum machining parameter combinations for obtaining higher depth of groove and smaller height of recast layer is a challenging task in LMM due to the presence of a large number of process variables. There is no perfect combination of parameters which can simultaneously result in both the highest depth of groove and lowest height of recast layer. This paper presents an attempt to develop a strategy for predicting the optimum machining parameter setting for the generation of the maximum depth of groove with minimum height of recast layer. A feed forward back-propagation neural network has been developed to model the machining process. The model, after proper training, is capable of predicting the response parameters as a function of four different control parameters. Experimental results demonstrate that the machining model is suitable and the optimization strategy satisfies practical requirements. The developed model has been found to be quite unique, powerful and flexible.     

Design and performance evaluation of a distribution network: a simulation approach

This paper introduces a scientific scheme for managing distribution networks to facilitate delivery of items from one place to another. It involves design of a distribution network, management of a fleet of trucks, and discussion of the roles of a distribution network. Initially, a distribution network was designed with an objective of minimizing total travelling time with the constraints of maximum truck capacity and maximum duration of travel for a single truck driver at one time. Simulation tests were then performed using SIMPROCESS to study the flow within the designed distribution network. Finally, the simulation results were used to assess the efficiency of the network, to determine the critical success factor of controlling the inventory level of packages in the distribution system, and to analyze the benefits of freight consolidation.     

利用排队模型优化零件机械加工时间

利用现代仿真技术,通过优化工件在各工序的加工时间,以消除工序中存在的工件等待加工现象。提出的方法还可用来预测机械制造系统的运行性能、比较各种可供选择的方案以及实现制造系统决策优化。     

交通信息采集无线传感器网络节点部署的微粒群优化方法

针对城市道路交通信息采集无线传感器网络节点部署优化问题,采用传感器网络连通性和覆盖性作为综合评价函数,以满足网络连通性和覆盖性为约束,建立节点部署的约束优化数学模型,并用罚函数法将其转化为无约束优化模型。采用微粒群算法求解,并用动态改变惯性权重方法作为改进算法解决微粒群算法的早熟收敛。以北京市二环以内的道路为例进行模拟实验,结果表明,微粒群算法及其改进算法使优化布局的评价函数值比初始手动布局提高1.71%和3.18%。微粒群算法及其改进算法能够优化交通信息采集的无线传感器网络节点布局。