带惯性权重的粒子群优化算法性能仿真

粒子群优化算法是一种随机搜索算法,并能以较大概率收敛到全局最优,微粒群算法中关键参数的选择方法对算法特性有显著影响.文中针对微粒群算法中的加速常数、惯性权重、取值范围、种群规模的设置对算法基本性能的影响进行了分析.实验结果证明:选择适合的参数设置水平,能够获得稳健和高效的优化效果.     

分区复杂性在时间序列分析中的应用

本文对KC复杂性,C₁、C₂复杂性等常见的时间序列分析中所用的方法进行了比较,在此基础上提出一种新算法——分区复杂性,同时将其应用在EEG时间序列分析、处理中,以检验它的有效性.     

单目标-多条件约束S-P网络可靠性优化的微粒群算法

研究2-状态单目标-多条件约束串-并联(S-P)网络的可靠性优化问题(RAP)。设计了具有压缩系数的离散型微粒群算法进行求解,采用Matlab编程对问题实例进行模拟仿真,结果表明,对于合理选择的初始解与算法参数,微粒群算法每次运行都收敛,并且能够收敛到最优解。通过与传统的智能算法(模拟退火算法、蚁群算法、遗传算法)比较,微粒群算法具有初始解容易选择、参数易于设置,收敛性好、收敛快的优势。     

改进的PSO算法在非线性模型参数辨识中的应用

参数辨识是过程建模的基础,对于参数辨识问题提出了许多不同的方法.针对传统模型参数辩识方法和遗传算法用于模型参数辨识时的缺点,提出一种基于微粒群优化(PSO)算法的模型参数辨识方法,利用PSO算法的强大优化能力,通过对算法的改进,将过程模型的每个参数作为微粒群体中的一个微粒,利用微粒群体在参数空间进行高效并行的搜索,以获得过程模型的最佳参数值,并将其用于对非线性系统模型的参数辨识,可有效提高参数辨识的精度和效率.该方法应用到实际例子中,获得了满意的辨识精度和效率,得到较为精确的过程模型,模型输出与实际输出基本一致,仿真结果令人满意.实例仿真结果表明,微粒群算法为非线性系统模型参数辨识提供了一种有效的途径.     

一种基于H∞滤波的模糊变 结构交互多模型算法

为提高对机动目标的跟踪效果,本文提出了一种基于H_∞滤波的模糊变结构交互多模型算法.该算法采用H_∞滤波器作为模型条件滤波器,通过调节其参数来增强对外界干扰的鲁棒性,提高滤波性能和精确度.同时把交互多模型(IMM)算法的模型概率作为模糊推理系统的输入,利用模糊逻辑推理优化模型概率,实时改变模型数目和参数,减少模型之间竞争,实现变结构功能.实验仿真结果表明,本文提出的算法比标准IMM算法具有更高的跟踪精确度.     

动态自适应微粒群优化算法

群体智能是目前智能领域非常活跃的新兴研究领域,微粒群算法作为其典型的实现形式,受到普遍的关注.本文分析了基本微粒群算法的特点,改善了动态自适应微粒群优化算法,实验结果证明该方法的优越性.     

通信

TN91 2005030642 求解通信优化问题的一种微粒群优化方法/原萍,王光兴,张洋洋(东北大学)//东北大学学报.-2004,25(10).-934-937 简述了微粒群优化算法的原理、流程及其参数,在此基础上,针对其在通信中的应用,提出了一种基于分布式计算的多目标微粒群算法——分割域多目标PSO算法,并将其用于基站优化问题,仿真研究结果表明, 它能很好地解决移动通信中的基站优化问题,并可被有效推广到处理诸如信道分配、网络拓扑优化设计,IP组播、Ad hoc簇结构及组播路由等通信服务.图4参10(木)     

群智能微粒群算法及参数分析

微粒群算法是继蚁群算法之后提出的又一种新型的进化计算技术,具有典型的群智能的特点。文中首先介绍了微粒群算法的基本原理、流程,在算法参数分析的基础上,运用测试函数对粒子在寻优过程中的轨迹的进行研究,有利于探讨微粒群算法的收敛性,全局和局部平衡性以及算法稳定性和复杂程度,对算法参数的选择提供有效的指导和借鉴意义。     

基于PSO的Kriging相关模型参数优化

Kriging插值计算过程中的相关模型参数确定是构造回归模型的关键,常用的模式搜索方法求解相关模型参数时存在计算精度依赖搜索起始点的缺点,从而导致最优解的不稳定.利用一种改进的二进制编码微粒群算法(Genetic Particle Swarm Optimization,GPSO)来求解相关函数的参数,该方法采用动态选择和调整变异算子概率的策略,克服了参数优化过程中对初始点设置的依赖问题,函数测试的性能比较表明该方法具有良好的收敛速度和稳定性.     

无线传感网络布局的虚拟力导向微粒群优化策略

无线传感网络通常由固定传感节点和少量移动传感节点构成,动态无线传感网络布局优化有利于提高无线传感网络覆盖率和目标检测概率,是无线传感网络研究的关键问题之一.传统的虚拟力算法在优化过程中容易受固定传感节点的影响,无法实现全局优化.本文结合虚拟力算法和微粒群算法,提出一种面向无线传感网络布局的虚拟力导向微粒群优化策略.该策略通过无线传感节点间的虚拟力影响微粒群算法的速度更新过程,指导微粒进化,加快算法收敛.实验表明,虚拟力导向微粒群优化策略能快速有效地实现无线传感节点布局优化.与微粒群算法和虚拟力算法相比,虚拟力导向微粒群优化策略不仅网络覆盖率高,且收敛速度快,耗时少.     

动态多目标优化的进化算法及其收敛性分析

给出了动态多目标优化问题的一种新解法.首先对时间变量进行了等区间离散化,在得到的子区间(称为环境)上定义了种群的静态序值方差和静态密度方差.然后把动态多目标优化问题近似地转化成了若干个两个目标的静态优化问题.在给出的一种能自动检测环境变化的应答算子下,提出了一种动态多目标进化算法,同时证明了算法的收敛性.计算机仿真表明新算法对动态多目标优化问题是有效的.     

Adaptive reduced-rank LCMV beamforming algorithms based on joint iterative optimization of filters: Design and analysis

This paper presents reduced-rank linearly constrained minimum variance (LCMV) beamforming algorithms based on joint iterative optimization of filters. The proposed reduced-rank scheme is based on a constrained joint iterative optimization of filters according to the minimum variance criterion. The proposed optimization procedure adjusts the parameters of a projection matrix and an adaptive reduced-rank filter that operates at the output of the bank of filters. We describe LCMV expressions for the design of the projection matrix and the reduced-rank filter. We then describe stochastic gradient and develop recursive least-squares adaptive algorithms for their efficient implementation along with automatic rank selection techniques. An analysis of the stability and the convergence properties of the proposed algorithms is presented and semi-analytical expressions are derived for predicting their mean squared error (MSE) performance. Simulations for a beamforming application show that the proposed scheme and algorithms outperform in convergence and tracking the existing full-rank and reduced-rank algorithms while requiring comparable complexity.     

一种协同神经网络不平衡参数的差分优化方法

协同模式识别方法是协同学原理应用于模式识别领域的一种新方法,识别过程中通过调整神经网络的注意参数,能改进系统的识别性能,对协同神经网络参数在不平衡条件下进行优化,能充分利用协同神经网络的自学习能力,以提高识别效果。差分进化作为一种有效的全局近似最优解的搜索算法,具有收敛性好、速度快的特点,文中提出了一种基于差分进化的方法对协同神经网络中参数进行优化,在协同神经网络的参数空间搜索最优参数,采用了均方适应度方差的机制自适应调整搜索速度和搜索精度,克服差分进化算法参数调整困难的不足,以提高算法的寻优能力,新方法具有全局兼局部寻优能力,不易陷入局部极值,同时新方法采用约简的序参量进化参数,使优化算法能有效提高协同神经网结的效率,实际图像的分类识别结果表明,注意参数的变化会导致完全不同的识别结果,另外,本文还将新算法与平衡参数的方法、其它优化的非平衡参数的协同学习算法进行了全局优化能力的比较,采用新方法具有更快的收敛速度和更优的分类识别效果。      

基于QGA的高压断路器数据融合方法的研究

为了消除传感器测量中的干扰及不确定性,获得更准确、更可靠的测量结果,提出一种新型的基于量子遗传算法的数据级融合算法,并将其应用于高压断路器机械振动或分合闸线圈电流等测量数据的融合。该算法以各传感器的加权因子为优化变量,以总的样本方差最小为目标函数,采用量子遗传算法进行优化求解,求得融合后的最优输出。通过各种比较,证明了该算法的正确性和实用性。该算法可推广应用于其他设备的数据融合。     

基于粒子群算法的激光位移传感器参量优化

为了提高激光三角测距精度、优化传感器设计过程中的光学参量,采用数学建模和计算机辅助分析的方法,解析了激光三角测距传感器中的关键参量与测量系统各指标的关系,并采用一种基于粒子群算法的参量优化方法,得到符合系统优化要求的光学参量,进行了理论分析和验证。结果表明,在进行参量设计时,各参量相互牵制;确定了粒子群搜索空间和约束;在灵敏度S_min达到2.2386mm时,系统分辨力可达到2.8μm,且其它各参量取值符合系统要求,同时优化效率大大提高。该优化方法算法简单、操作方便。     

Analog circuit optimization using adjoint network based sensitivity analysis

Performance optimization as per the desired specifications is a major requirement of analog and mixed signal circuit design process. Rapid scaling of the semiconductor technology demands efficient optimization techniques with minimal manual efforts. In this paper, a gradient based method for analog circuit optimization using adjoint network based sensitivity analysis is presented. The sensitivity of circuit response with respect to the different parameters is computed by using analog circuit and its adjoint transformation. The proposed method is applied to optimize performance of a two stage operational amplifier (OpAmp). Subsequently, the OpAmp circuit is simulated using Cadence Virtuoso for optimized parameters and the results are validated with post fabrication measurement results.     

Construction of optimal subband coders using optimized and optimalquantizers

A method is presented for the optimization of arbitrary quantizers by use of a compensating postfilter. It is shown that the resulting optimized quantizers fit the model of a linear time-invariant filter followed by additive noise uncorrelated with the input which also characterizes the optimal (Lloyd-Max) quantizers. On the basis of this model, an expression for the variance of the error of a subband coder using optimized quantizers is explicitly determined. Given analysis filters which statistically separate the subbands, it is shown that this variance is minimized if these synthesis filters are chosen, which would achieve perfect reconstruction in lossless coding. The globally optimum filter bank, minimizing the coder error variance, is further obtained by proper choice of its analysis filters. A novel method for the determination of optimal bit allocation to subbands of the filter banks with optimized quantizers is also developed. The results are evaluated experimentally by comparison of the optimum uniformly split subband image coding scheme to classical logarithmically-split filter bank (wavelet) coding methods.     

Blind adaptive algorithms for minimum variance CDMA receivers

Constrained optimization methods have received considerable attention as a means to derive blind multiuser receivers with low complexity. The receiver's output variance is minimized subject to appropriate constraints which depend on the multipath structure of the signal of interest. When multipath is present, the constraint equations can be written in parametric form, and the constraint parameters jointly optimized with the linear receiver's parameters. We develop adaptive solutions for this joint, constrained optimization problem. Both stochastic gradient and recursive least-square-type algorithms are developed. The performance of the proposed methods is compared with other blind and trained methods and turns out to be close to the trained minimum mean-square-error receiver     

Iterative-based visualization-oriented fusion scheme for hyperspectral images

This article investigates a novel visualization-based fusion of hyperspectral image bands using an iterative approach. Given a multi-objective function and the pixel-based hyperspectral image fusion method, the optimization process is described as finding the optimal fusion parameters to improve the fusion performance. Accordingly, an iterative-based approach is adopted. In the first step, the fusion process is developed using the pixel-based fusion technique. In the second step, the fused image is produced, and the fusion quality is assessed for multi-objective function construction. For multi-objective formulation, we focus three desired properties of the fused image such as entropy, variance, and smoothness. In the last step, fusion parameters are updated iteratively by examining the objective function. Here, the self-adaptive learning particle swarm optimizer is used to refine the fusion parameters iteratively. Different hyperspectral images, such as Cuprite mining, AVIRIS Indian pines scene, are employed in the evaluation. Quantitative analysis of fused images is carried out through some efficient fusion metrics such as correlation coefficient, entropy, Q-average, ERGAS, SAM, and SID. Experimental results show that the proposed approach outperforms existing methods in terms of both objective function criteria and visual effect.     

方差分析在等离子表面冶金中的应用

方差分析是根据试验的结果进行分析,鉴别各个有关因素对试验结果影响的有效方法.将应用方差分析研究主要工艺参数对Ti2AlNb金属间化合物等离子表面渗Mo显微硬度的影响.单因素方差分析试验结果表明:工艺参数如工作温度、保温时间、极间距对Ti2AlNb金属间化合物等离子表面渗Mo显微硬度均有显著影响,而工作气压对Ti2AlNb金属间化合物等离子表面渗Mo显微硬度无显著影响.