Adaptive Multi-layer Particle Swarm Optimization with Neighborhood Search

Particle swarm optimization (PSO) has shown a good performance on solving global optimization problems.Traditional PSO has two main drawbacks of premature convergence and low convergence speed,especially on complex problems.This paper presents a new approach called Adaptive multi-layer particle swarm optimization with neighborhood search (AMPSONS),where the traditional PSO is improved by employing an adaptive multi-layer search and neighborhood search strategy to achieve a trade-off between exploitation and exploration abilities.In order to evaluate the performance of the proposed AMPSONS algorithm,the performance of AMPSONS is compared with five other PSO family algorithms,namely,CLPSO,DNLPSO,DNSPSO,global MLPSO and local MLPSO on a set of benchmark functions.The comparison results show that AMPSONS has a promising performance on majority of the test functions.     

一种改进粒子群优化算法

作为群集智能的代表性方法之一，粒子群优化(PSO)算法通过粒子之间的合作与竞争以实现对多维复杂空间的高效搜索。提出了一种改进粒子群优化(MPSO)算法。MPSO同时采用局部模式压缩因子方法和全局模式惯性权重方法以获得相对较高的性能。针对PSO算法可能出现的停滞现象，MPSO引入了基于全局信息反馈的重新初始化机制。数值仿真结果显示了该算法的有效性。     

A Modified Particle Swarm Optimizer and its Application to the Design of Microwave Filters

In this paper, a modified particle swarm optimization (PSO) algorithm is presented and its applicability is shown for the design of specific microwave filter as a case study of microwave components. In the proposed modified PSO algorithm, particles in the swarm are divided to form multiple sub-swarms. The social component of PSO’s velocity update equation is modified to include the effects of multiple sub-swarms. Five benchmark functions have been considered for testing the proposed algorithm. The approach has been tested for two basic modifications of PSO namely PSO with inertia weight (IW) and PSO with constriction factor method (CFM). The simulated results illustrate that the modified PSO algorithm has the potential to converge faster, thus reducing the computational expenses, while maintaining/improving the quality of solution. Finally, the proposed algorithm is used for the design of coupled microstrip line band pass filter which is a computationally expensive process when the design is conducted using evolutionary algorithms and electromagnetic (EM) simulation tools.     

改进粒子群算法的露天矿路径优化研究

为将面向连续优化的粒子群优化算法应用于露天矿路径优化问题的求解,提出了露天矿路径优化问题的权重编码方案．该方案将属于组合优化的露天矿路径问题转化为连续优化问题,同时保留了粒子群算法的易操作性和高效性．针对粒子群算法易陷入局部最优的问题,提出了适合露天矿路径优化问题的基于k-中心点法的改进措施,在此基础上采用k-中心点法对粒子群进行聚类分析,实现了粒子之间的信息交换,扩大了粒子的搜索空间,避免算法陷入局部最优．最后,用露天矿的实际路径节点数据验证了权重编码方案和改进粒子群算法的有效性．     

Pattern synthesis of centre fed linear array using Taylor one parameter distribution and restricted search Particle Swarm Optimization

In this article, a new method of pattern synthesis of centre fed, equal distance linear array having single and multiple synthesis objectives has been proposed and statistically investigated. Single objective of reduced side lobe level (SLL) and first null beamwidth (FNBW) has been considered separately. Consequently, multiple objectives of beamwidth and side lobe level have been investigated. Synthesis of linear array for suitable objectives has been investigated on Taylor one parameter distribution with equal progressive phase. Excitation amplitude of each array element is taken as optimization parameter where distribution has been optimized using Particle Swarm Optimization (PSO) for achieving low SLL. Later the same has been incorporated for obtaining suitable FNBW. In our optimization algorithm conventional PSO has been modified with a restricted search PSO (RSPSO) where search space has been predefined within excitation amplitude range. PSO within the defined range searches for optimum excitation amplitude to achieve the desired objectives. In order to illustrate the effectiveness of the proposed RSPSO, simulation results of three significant instances of linear array have been presented for both even and odd number of element. The design results obtained using RSPSO have improved result than those obtained using other state of the art evolutionary algorithms like differential evolution (DE), invasive weeds optimization (IWO) and Conventional particle Swarm optimization (CPSO) in a statistically significant way.     

基于改进BP神经网络和粒子群优化算法的图像滤波方法的研究

提出了一种基于改进BP神经网络和粒子群优化算法(PSO)的图像滤波方法.该方法利用双曲正切形式的误差函数代替BP神经网络传统的最小均方误差函数(LMS),并将改进后的BP神经网络利用PSO算法优化,用来减小图像噪声对神经网络精度的影响以及避免神经网络陷入局部极小值点,从而提高神经网络去噪能力.实验结果表明,与传统滤波方法相比,该方法不仅能有效地滤除图像中的高斯噪声而且能很好地保护图像细节.     

A modified PSO technique using heterogeneous boundary conditions for broadband compact microstrip antenna designing

A modified particle swarm optimization (PSO) variant is implemented on a conventionally optimized broadband antenna. Proposed algorithm uses heterogeneous boundary conditions to particles that fly out of the search space. Search space boundary conditions were chosen depending on which bound was violated. The algorithm is executed using MATLAB for the PSO computations and IE3D. Microstrip antennae inherently suffer from low bandwidth. Thus, a broadband patch antenna is optimized to improve the antenna return loss bandwidth in order to validate the proposed algorithm. For the patch antenna, a significant 12 % bandwidth improvement and 20.84 % size reduction was achieved. It was fabricated and satisfactory conformity was found with simulated results.     

参数自适应混沌粒子群算法在盲源分离中的应用

独立分量分析(ICA)是盲源信号分离中应用最为广泛技术,其应用过程需要对目标函数进行优化,传统粒子算法(PSO)对其进行优化时,存在易陷入局部最优、稳定性差等缺陷,针对此问题,提出采用参数自适应混沌粒子群算法对ICA进行优化.首先采用对PSO的参数进行自适应调整,提高粒子的搜索能力,然后对粒子群进行混沌扰动,提高算法收敛速度.仿真结果表明,使用参数自适应混沌粒子群算法可以有效解决ICA的目标函数优化问题,极大提高了盲源信号的分离效果.     

Detection of Cardiac Signal Characteristic Point Using Log-Domain Wavelet Transform Circuits

To meet the requirement of low power consumption in biomedical implantable pacemaker applications, a novel method based on balanced log-domain wavelet transform (WT) circuits has been developed for detecting QRS complexes of cardiac signals. By using a hybrid particle swarm optimization algorithm (PSO) combined with sequential quadratic programming, an excellent approximation of the first derivative of a Gaussian wavelet is achieved. The WT circuits are composed of filters whose impulse response is the approximation of the Gaussian wavelet. The WT filter design is based on a time inverse follow-the-leader feedback structure with class-AB balanced log-domain integrators as the main building blocks. HSPICE simulation shows that the power consumption is only 62 nW per scale for a 1.2 V supply, and the dynamic range is 86 dB for 2% total harmonic distortion. The high accuracy of the QRS complex detection method has been validated using the MIT-BIH database. This work was supported by the National Natural Science Foundation of China under Grant No. 50677014, Doctoral Special Fund of Ministry of Education under Grant No. 20060532002, High-Tech Research and Development Program of China (No. 2006AA04A104), the Program for New Century Excellent Talents in University of China (NCET-04-0767), and the Foundation of Hunan Provincial Science and Technology (06JJ2024).     

基于微粒群的矢量量化码书设计算法

针对在LBG算法中存在初始码书的选择极易影响码书训练的收敛速度和最终码书性能的缺陷,提出了一种基于微粒群的矢量量化码书设计算法.首先产生具有一定全局性特点的初始码书,然后再应用LBG算法进行优化得到同时具有局部特性的码书.实验结果验证了该算法的合理性.     

Parallel particle swarm optimization and finite- difference time-domain (PSO/FDTD) algorithm for multiband and wide-band patch antenna designs

This paper presents a novel evolutionary optimization methodology for multiband and wide-band patch antenna designs. The particle swarm optimization (PSO) and the finite-difference time-domain (FDTD) are combined to achieve the optimum antenna satisfying a certain design criterion. The antenna geometric parameters are extracted to be optimized by PSO, and a fitness function is evaluated by FDTD simulations to represent the performance of each candidate design. The optimization process is implemented on parallel clusters to reduce the computational time introduced by full-wave analysis. Two examples are investigated in the paper: first, the design of rectangular patch antennas is presented as a test of the parallel PSO/FDTD algorithm. The optimizer is then applied to design E-shaped patch antennas. It is observed that by using different fitness functions, both dual-frequency and wide-band antennas with desired performance are obtained by the optimization. The optimized E-shaped patch antennas are analyzed, fabricated, and measured to validate the robustness of the algorithm. The measured less than - 18 dB return loss (for dual-frequency antenna) and 30.5% bandwidth (for wide-band antenna) exhibit the prospect of the parallel PSO/FDTD algorithm in practical patch antenna designs.     

Dielectric absorber design for wide band–wide oblique incidence angle using genetic algorithm

With an aim to design a V-shape absorber using a global optimization algorithm, the reflectivity value of the absorber is calculated as a cost function by using an equivalent multilayered absorber model, and is minimized by the optimization algorithm. Both normal and oblique incidences of plane wave are analyzed. The optimization parameters are the relative permittivity and the length of the multilayered absorber over a wide frequency band and wide oblique incidence angle of a plane wave. The optimized results for different frequency bands and incidence angles show that the minimized reflection coefficients and corresponding relative permittivity are within an acceptable range for each application considered in this work.     

区域分割的自适应变异粒子群算法

为了提高粒子群算法（PSO）的收敛性及多样性,提出一种基于区域分割的自适应变异粒子群算法（RSVPSO）.算法采用区域分割的思想,利用粒子间信息交叉,使粒子搜索区间快速缩小;同时在迭代后期与自适应变异策略相结合,提高粒子跳出局部最优陷阱的能力和增强粒子多样性,达到寻优的目的.将所提出的算法应用于8个测试函数,并与精英免疫克隆选择的协同进化粒子群等算法进行比较,结果表明,新算法在收敛速度、搜索精度及寻优效率等方面有较大提高.     

混合粒子群和差分进化的定位算法

针对传统无线传感器网络(wireless sensor network,WSN)中节点定位精度不高的问题,提出了一种混合粒子群(particle swarm optimization,PSO)和差分进化优化(differential evolution,DE)算法。首先在PSO中引入惯性权重的自适应更新策略,以兼顾开发和勘探能力,在种群经过PSO进化后,然后根据提前设定的阈值,将其分为适应度值较大的Su种群和适应度值较小的In种群,In中的粒子使用DE算法继续优化。HPSO-DE算法结合PSO算法和DE算法的优点,达到较好的性能。然后用标准测试函数来检测该算法的性能,验证结果表明所提出的HPSO-DE在寻优速度和收敛精度较PSO和DE而言都有了较大提高。接下来将HPSO-DE方法应用到WSN网络节点定位场景上,从实验测试结果可以看出,其精度相比PSO平均提高了0.5 m左右,在定位上具有更大的优势。     

基于人工鱼群算法的IIR数字滤波器设计

将人工鱼群算法（AFSA）用于IIR数字滤波器设计，建立了相应的优化模型，给出了简化的人工鱼群算法及其实现步骤。最后，将该算法用于低通、带通IIR数字滤波器的设计，并与粒子群算法进行了比较。仿真结果证明了AFSA的有效性，并且具有算法灵活、简单，全局收敛性好。收敛速度快的优点。     

Introducing the fractional-order Darwinian PSO

One of the most well-known bio-inspired algorithms used in optimization problems is the particle swarm optimization (PSO), which basically consists on a machine-learning technique loosely inspired by birds flocking in search of food. More specifically, it consists of a number of particles that collectively move on the search space in search of the global optimum. The Darwinian particle swarm optimization (DPSO) is an evolutionary algorithm that extends the PSO using natural selection, or survival of the fittest, to enhance the ability to escape from local optima. This paper firstly presents a survey on PSO algorithms mainly focusing on the DPSO. Afterward, a method for controlling the convergence rate of the DPSO using fractional calculus (FC) concepts is proposed. The fractional-order optimization algorithm, denoted as FO-DPSO, is tested using several well-known functions, and the relationship between the fractional-order velocity and the convergence of the algorithm is observed. Moreover, experimental results show that the FO-DPSO significantly outperforms the previously presented FO-PSO.     

改进的多目标粒子群算法优化设计及应用

针对粒子群算法存在易陷入局部最优点的缺点,提出了一种改进的带变异算子的多目标粒子群优化算法。采用非支配排序策略和动态加权法选择最优粒子,引导种群飞行,提高帕累托(Pareto)最优解的多样性。与其他优化算法相比,该算法易于实现并且计算速度更快。通过计算Pareto前沿最优解设计最佳多层电磁吸收体,在吸收体的厚度与反射系数之间取得最佳折衷。通过对反射系数函数与吸收体厚度函数测试验证,该算法能够在保持优化解多样性的同时具有较好的收敛性。     

A New Class of Hybrid Particle Swarm Optimization Algorithm

A new class of hybrid particle swarm optimization （PSO） algorithm is developed for solving the premature convergence caused by some particles in standard PSO fall into stagnation. In this algorithm, the linearly decreasing inertia weight technique （LDIW） and the mutative scale chaos optimization algorithm （MSCOA） are combined with standard PSO, which are used to balance the global and local exploration abilities and enhance the local searching abilities, respectively. In order to evaluate the performance of the new method, three benchmark functions are used. The simulation results confirm the proposed algorithm can greatly enhance the searching ability and effectively improve the premature convergence.     

Design of two-dimensional IIR digital filters by using a novel hybrid optimization algorithm

This paper proposes a hybrid optimization algorithm named as BBO–PSO, which is a combination of biogeography-based optimization (BBO) and particle swarm optimization (PSO). In BBO–PSO, the whole population will be split into several subgroups and BBO is employed for local search in each subgroup independently to achieve the different local optima while PSO is employed for global search based on the local optima to achieve the global optimum. The test results on the benchmark functions show that BBO–PSO has powerful search ability with great robustness. Furthermore, the proposed algorithm is applied to the design of the 2-D IIR digital filters and the simulation results show that it outperforms the existing methods on this problem.     

An intelligent method to design laser resonator with particle swarm optimization algorithm

In order to design a complex laser resonator with multi-parameters, the method of particle swarm optimization (PSO) algorithm is employed. The parameters influencing the resonator stability and mode size distribution are taken into consideration, and the stability criteria index and the mode size distribution are used as target values. The absolute values of the differences between practical and the target values are set as the fitness function for the PSO. By minimizing the fitness function, a laser resonator with the optimized cavity parameters can be found. The analyses for the design example demonstrate the feasibility and validity of the PSO method in the computer aided design of multi- parameters laser resonator. Applying PSO algorithm in the intelligent design of solid state laser resonators can realize the transition from manual trial-and-error to computer intelligent design of the laser resonators.