基于空间缩放和吸引子的粒子群优化算法

为解决粒子群优化(Particle Swarm Optimization,PSO)算法中粒子越界、算法进化后期收敛速度慢和早熟收敛的问题,通过分析PSO算法中粒子运动行为和算法稳定性,提出了一种基于空间缩放和吸引子的粒子群优化(PSO with search space zoomed factor and attrac...     

进化粒子群优化算法及其在各向异性含水层参数估计中的应用

在粒子群算法的基础上,引入进化思想和组群思想提出一种新的智能优化算法———进化粒子群算法( EPSO)。基于抽水试验数据,将EPSO算法应用到各向异性含水层参数估计中,对算法性能进行研究并与其他算法进行了对比,发现标准粒子群算法及其一般改进算法已不能有效求解各向异性含水层参数,而EPSO算法进行多次计算后,1)结果可靠;2)目标函数值及待估参数稳定;3)对初始范围的鲁棒性好。结果表明EPSO算法对各向异性含水层参数估计问题具有可靠性、收敛性和鲁棒性,可望应用到更广泛的参数识别问题中。     

粒子群算法求解连续优化问题的性能分析

粒子群优化(PSO)算法是一类有效的随机全局优化技术,适用于求解连续优化问题.它利用一个粒子群搜索解空间,通过粒子间的相互作用发现复杂搜索空间中的最优区域.本文介绍了基本的PSO算法,使用3类代表性的标准测试函数对粒子群算法进行了实验分析,并进一步讨论了PSO算法的寻优性能,提出了PSO求解连续优化问题的性能分析策略.     

基于莱维飞行的粒子群优化算法

为了有效解决粒子群优化算法易陷入局部最优的缺陷,在粒子群优化优化算法（PSO）的基础上,引入莱维飞行,提出了一种基于莱维飞行的粒子群优化算法（LPSO）。该算法在迭代过程中,对粒子位置进化效果进行判断,若粒子多次迭代后仍无法进化到更优位置,则使用莱维飞行更新粒子位置。改进后的算法增加了粒子位置变化的活力,提高了算法的有效性。仿真实验结果表明,该算法在求解全局最优时,效果优于原始粒子群优化算法,在多峰值函数优化问题中其优越性更加突出。     

多向学习自适应的粒子群算法

粒子群优化算法（PSO）是一种群体智能算法,通过粒子间的竞争和协作以实现在复杂搜索空间中寻找全局最优点。但基本PSO算法存在进化后期收敛速度慢、易陷入局部最优点的缺点,提出了一种多向学习型的粒子群优化算法,该算法中粒子通过同时追随自己找到的最优解、随机的其他粒子同维度的最优解和整个群的最优解来完成速度更新,通过判别区域边界来完成位置优化更新,通过对全局最优位置进行小范围扰动,以增强算法跳出局部最优的能力。对几种典型函数的测试结果表明：改进后的粒子群算法明显改善了全局搜索能力,并且能够有效避免早熟收敛问题。算法使高维优化问题中全局最优解相对搜索空间位置的鲁棒性得到了明显提高,适合于求解同类问题,计算结果能满足实际工程的要求。     

一种基于差异演化变异的粒子群优化算法

为了保持粒子种群的多样性而避免发生“早熟”的问题,提出一种基于差异演化变异的粒子群优化算法（PSO）,该方法通过粒子聚集性判断如果粒子群中的粒子过于聚集,则使用差异演化算法对PSO算法中各个粒子的自身历史最佳位置进行变异,以实现保持粒子群种群多样性的目的。对4种常用函数的优化问题进行测试并进行比较,结果表明：所改进的粒子群优化算法比标准粒子群优化算法更容易找到全局最优解,优化效率和优化性能明显提高。     

基于文化框架的随机粒子群优化算法

提出了随机粒子群优化算法(rPSO),并将其与标准PSO纳入到文化算法(CA)框架中,建立了基于文化框架的随机粒子群优化算法(CA-rPSO)。该算法以rPSO作为信念空间的进化算法,以PSO作为群体空间的进化算法,形成了两者独立并行进化的"双演化双促进"机制。选取5个测试函数进行了仿真实验分析并与其他算法进行了比较,结果表明CA-rPSO的寻优性能得到显著提高,且算法简单、易于实现。     

基于差分演化的粒子群算法

粒子群优化算法是一种简单有效的随机全局优化算法.但粒子群优化算法有易陷入局部极值点,进化后期收敛速度慢,精度较差的缺点.为了改进粒子群优化算法,将差分演化算法融合到粒子群优化算法中,在算法中,将粒子每代的所有局部最优位置进行变异、杂交、选择操作,提出了基于差分演化的粒子群算法.使粒子群算法和差分演化的探测和开发能力得到有效利用与平衡,提高了求解进度和效率,并通过仿真验证算法的性能优于带线性递减权重的粒子群优化算法和差分演化算法.     

适应性粒子群寻优算法Ⅱ

适应性粒子群寻优算法Ⅰ(APSO-Ⅰ)是在有序的决策中始终引入随机的、不可预测的决定.为解决APSO-Ⅰ算法收敛深度不够的问题,提出适应性粒子群寻优第Ⅱ代算法(APSO-Ⅱ).APSO-ⅡⅢ算法是将有序(标准PSO粒子群寻优)和无序(自适应寻优)进行适当的分离.以发挥各自的优势.在自适应寻优阶段,通过在最优柱子邻域空间探寻更优化的解.一但新的优化解被发掘,便利用标准PSO快速寻优.典型复杂函数优化的仿真结果表明,APSO-Ⅱ在收敛速度和收敛深度上均优于DPSO(耗散型PSO),HPSO(自适应层次PSO),EPSO(自适应逃逸PSO)和APSO-I.     

一种群活性反馈粒子群优化算法

为解决粒子群优化(PSO)算法的早熟收敛问题,提出一种群活性反馈PSO进化算法SAF-PSO。利用群活性加速度作为多样性测度,当群活性加速下降时,对粒子的位置和速度分别执行进化和变异操作,增强粒子跳出局部最优的能力,提高寻找全局最优的几率。对基准函数的仿真结果表明,与其他PSO算法相比,该算法具有更强的全局搜索能力和更高的寻优精度。     

增强型微粒群优化算法及其在软测量中的应用

对微粒群优化算法(PSO)进行分析,提出一种增强型微粒群优化算法(EPSO),用EPSO和PSO对几种常用函数的优化问题进行测试比较,结果表明EPSO比PSO更容易找到全局最优解,优化效率和优化性能明显提高,将EPSO用于催化裂化装置主分馏塔粗汽油干点软测量,建立了基于EPSO算法的粗汽油干点神经网络软测量模型,研究结果表明,基于EPSONN的软测量模型比基于BPNN的软测量模型具有更高的精度和更好的性能。     

Active target particle swarm optimization

We propose an active target particle swarm optimization (APSO). APSO uses a new three‐target velocity updating formula, i.e. the best previous position, the global best position and a new target position (called active target). In this study, we distinguish APSO from EPSO (extended PSO)/PSOPC (PSO with passive congregation) by the different methods of getting the active target. Note that here EPSO and PSOPC are the two existing methods for using three‐target velocity updating formula, and getting the third (active) target from the obtained positions by the swarm. The proposed APSO gets the active (third) target using complex method, where the active target does not belong to the existing positions. We find that the APSO has the advantages of jumping out of the local optimum and keeping diversity; however, it also has the disadvantages of adding some extra computation expenses. The experimental results show the competitive performance of APSO when compared with PSO, EPSO, and PSOPC. Copyright © 2007 John Wiley & Sons, Ltd.     

一种增强型的粒子群优化算法

针对粒子群优化算法在进化后期容易陷入局部最优的缺点,提出了一种增强型的粒子群优化算法,即当粒子陷入局部极值点时,从增强粒子的自我学习能力,增强种群中其他相关粒子探索新区域的能力和增强粒子之间的信息交流三个方面来增强算法的寻优能力。数值实验结果表明,新算法具有很好的寻优性能。     

基于EPSO-BP 的Elman 网络及其在飞行轨迹预测中的应用

针对BP 训练方式采用梯度法易导致局部收敛的不足, 提出一种融合进食粒子群算法(EPSO) 和梯度法的Elman 网络优化方法. 首先, 通过模拟鸟群进食行为得到一种EPSO 算法, 以改善标准PSO 的全局性能; 然后, 将EPSO 用于Elman 网络权值的全局优化, 同时将梯度法用于EPSO 的进食过程局部搜索, 以提高解的局部收敛性能; 最后, 将该网络优化方法用于飞行轨迹预测实验, 仿真结果表明了其有效性.

     

A new algorithm for flexible job-shop scheduling problem based on particle swarm optimization

This paper proposes a new algorithm, named EPSO, for solving flexible job-shop scheduling problem (FJSP) based on particle swarm optimization (PSO). EPSO includes two sets of features for expanding the solution space of FJSP and avoiding premature convergence to local optimum. These two sets are as follows: (I) particle life cycle that consists of four features: (1) courting call—increasing the number of more effective offspring (new solutions), (2) egg-laying stimulation—increasing the number of offspring from the better parents (current solutions), (3) biparental reproduction—increasing the diversity of the next generation (iteration) of solutions, and (4) population turnover—succeeding the population (the current set of all solutions) in the previous generation by a population in a new generation that is as able but more diverse than the previous one; and (II) discrete position update mechanism—moving particles (solutions) towards the flight leader (the best solution), namely, interchanging some integers in every solution with those in both the best solution and itself, using similar swarming strategy as the update procedure of the continuous PSO. The basic objective function used was to minimize makespan which is the most important objective, hence, providing the simplest way to measure the effectiveness of the generated solutions. Benchmarking EPSO with 20 well-known benchmark instances against two widely-reported optimization methods demonstrated that it performed either equally well or better than the other two.     

An extended particle swarm optimization algorithm for pattern synthesis of conformal phased arrays

In this article, an extended particle swarm optimization (EPSO) algorithm is proposed for designing conformal phased arrays. On the basis of traditional particle swarm optimization (PSO), novel velocity updating mechanism, new exceeding boundary control operator, and global best perturbation are introduced in EPSO to overcome the drawbacks of PSO. To validate the efficiency of the proposed algorithm, both the classical test functions and the scenarios concerning a 1 × 9‐element cylindrical conformal phased array and a 3 × 9(27)‐element cylindrical conformal array with flat‐top shaped‐beam pattern are presented. Simulation results show that the proposed method is superior to genetic algorithm (GA) and PSO when applied to both the classical test functions and the practical problems of conformal antenna array synthesis. © 2010 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2010.     

改进的多种群协同进化微粒群优化算法

提出一种改进的基于多种群协同进化的微粒群优化算法（ＰＳＯ）．该算法首先利用免疫算法实现解空间的均匀划分,增加了算法稳定性和全局搜索能力．在运行过程中,通过种群进化信息生成解优胜区域,指导变异生成的微粒群向最优解子空间逼近,提高算法逃出局部最优的能力．将此算法与ＰＳＯ 算法和多种群协同进化微粒群算法进行比较,数据实验证明,该算法不仅能有效地克服其他算法易陷入局部极小值的缺点,而且全局收敛能力和稳定性均有显著提高．     

Unit commitment problem using enhanced particle swarm optimization algorithm

This paper proposes an enhanced PSO (EPSO) approach to solve the unit commitment (UC) problem in electric power system, which is an integrated improved discrete binary particle swarm optimization (DBPSO) with the Lambda-iteration method. The EPSO is enhanced by priority list based on the unit characteristics and heuristic search strategies to repair the spinning reserve and minimum up/down time constraints. The implementation of EPSO for UC problem consists of three stages. First, the DBPSO based on priority list is applied for unit scheduling when neglecting the minimum up/down time constraints. Second, heuristic search strategies are used to handle the minimum up/down time constraints and decommit excess spinning reserve units. Finally, Lambda-iteration method is adopted to solve economic load dispatch based on the obtained unit schedule. To verify the advantages of the EPSO method, the EPSO is tested and compared to the other methods on the systems with the number of units in the range of 10 to 100. Numerical results demonstrate that the EPSO is superior to other methods reported in the literature in terms of lower production cost and shorter computational time.     

嵌入极值优化的混合粒子群优化算法

针对标准粒子群算法容易陷入局部极值和精度低的问题,提出一种嵌入极值优化算法的粒子群优化算法。在线性下降的惯性权重粒子群算法运行过程中,间隔一定迭代次数与极值优化算法相结合,利用其波动性增加种群的多样性,并有效结合粒子群算法较强的全局探索能力和极值优化算法精细的局部搜索性能,以较高精度收敛到全局极值。仿真实验结果表明,该混合算法是一种求解高维多峰连续函数极值的有效方法。     

一种协调勘探和开采能力的粒子群算法

提出一种新的协调勘探和开采能力的粒子群优化算法. 该算法将种群分为随机子群和进化子群, 随机子群增加了算法全局解空间的勘探能力, 在运行过程中通过随机子群进化信息生成解优胜区域指导进化粒子向着最优解子空间逼近. 为了提高算法收敛速度, 算法只在进化子群进入收敛阶段时才对其进行指导, 以防止增加种群多样性导致算法开采能力下降的问题. 将此算法与其他改进粒子群算法进行比较, 实验结果表明, 该算法有较好的全局收敛性, 不仅能有效地克服其他算法易陷入局部极小值的缺点, 而且算法收敛速度和稳定性都有显著提高.