粒子群优化算法的研究

粒子群优化算法是一种基于群体智能的优化算法。介绍粒子群优化算法及其基本原理,给出各种改进技术及研究现状,并展望未来的发展趋势。     

混合粒子群优化算法及其应用

提出了一种通过改进全局最优位置粒子寻优策略而提高粒子群优化计算效率的混合粒子群优化算法。针对流程工业典型设备的状态跟踪预报等有计算时间限制的优化问题,混合粒子群优化算法在不改变原有粒子群粒子寻优策略的前提下,将粒子群整体已搜寻到的全局最优位置看作一个特殊的粒子,令该粒子执行梯度下降寻优的寻优策略。在粒子群的寻优迭代计算中增加全局最优位置粒子单独的梯度下降寻优过程,从而将粒子群优化算法的全局寻优特性与梯度下降算法的邻域寻优特性相结合,以提高粒子群优化算法的整体寻优效率,进而缩短寻优计算的时间。针对流程工业典型设备的实际应用表明,混合粒子群优化算法能够减少寻优迭代次数,进而缩短优化计算时间。     

改进的粒子群优化算法及其在多目标过程设计中的应用

介绍了粒子群优化算法的基本思想和步骤,对基于动态Pareto解集的多目标粒子群优化算法进行了分析,提出了随机更新个体粒子最优位置的策略,并把改进的粒子群优化算法用于实际的多目标过程设计中.对甲苯加氢脱烷基化过程进行了分析,利用此方法对以经济和环境为目标的加氢脱烷基化过程设计进行了计算.结果表明,此种方法可以用于实际的多目标过程设计中.     

一种求解图着色问题的改进粒子群算法

图着色问题是一个典型组合优化难题,文章尝试用改进的粒子群算法来解决此问题。为了增强粒子的搜索能力和提高粒子群算法的收敛速度,在传统的粒子群算法中引入了动态惯性权值递减的策略与记忆机制。实验表明,该算法在性能上明显优于传统的粒子群算法。     

混合粒子群算法求解单机批调度问题

设计了一种混合粒子群算法(Hybrid Particle Swarm Optimization,HPSO)以求解基于工件动态到达的最小化最大拖期时间单机批调度问题。该算法在标准粒子群算法的基础上引入了惯性权重正弦调整,以改善标准粒子群算法的收敛速度和全局收敛性,然后采用自适应变异全局极值算法增强粒子群优化算法跳出局部最优解的能力,防止算法陷入局部最优。应用改进的算法对实验设计问题进行求解,证明了改进算法的有效性。     

基于差分进化粒子群混合优化算法的软测量建模

针对乙烯生产过程中,用传统方法难以直接完成对乙烯收率的在线测量的问题,提出了一种新型差分进化粒子群混合优化算法,建立了乙烯收率软测量建模。改进算法将优化过程分成两阶段,两分群分别采用粒子群算法和差分进化算法同时进行。迭代过程中引入进化速度因子进行算法局部收敛性判断,通过两个群体间的信息交流阻止算法陷入局部最优。对高维复杂函数寻优测试表明,算法的整体优化性能均强于基本粒子群算法和差分进化算法。应用结果表明,基于改进算法的软测量模型具有测量精度较高、泛化性能较好等优点。     

基于改进粒子群-差分算法的电容器优化配置

提出了利用改进粒子群-差分算法对配电网中电容器的最合适补偿位置和最佳的补偿容量进行优化计算。为了改善粒子群优化算法在迭代后期易于陷入局部极值的缺点,对其关键参数进行改进,将改进后的粒子群算法和差分算法相结合提高了粒子的搜索精度和全局搜索能力。通过对IEEE33节点配电系统进行仿真,验证了改进粒子群-差分算法在配电网电容器优化配置过程中的有效性和实用性。     

基于改进粒子群算法的TTCAN系统矩阵的优化

在基于时间触发的CAN总线控制系统中,TTCAN协议的任务调度是一个很重要的问题。针对这一问题,提出了基于改进粒子群算法的TTCAN系统矩阵优化方法,给出了详细的优化过程,并对算法进行了仿真研究。结果表明,改进的粒子群算法有效地减少了CAN的干扰抖动问题,提高了TTCAN网络带宽利用率,使得在实时控制系统中总线负载达到峰值时仍然能得到可靠的传输。     

全局群搜索优化算法及其在汽油调合中的应用

汽油调合配比生产优化是一种非线性约束的多峰优化问题。针对一般群智能优化算法在解决此类优化中易陷于局部最优解,提出了一种改进的群搜索优化算法--全局群搜索优化算法（GGSO）。该算法采用混沌机制初始化粒子在解空间内均匀分布;在算法前期,保留GSO的追随者进化策略,以保证算法的收敛速度。在算法后期,对追随者引入速度更新和个体最优,以保证算法的收敛精度;在粒子陷入局部极值时,对追随者和游荡者引入一种新的交叉、变异机制和自适应混沌扰动机制,以保证粒子跳出局部极值,提高算法全局寻优性能。分别用4个标准测试函数对优化算法进行测试,结果表明：GGSO算法与标准GSO、线性递减惯性权重粒子群算法（LDWPSO）比较,收敛速度和全局寻优性能有明显优势。汽油在线调合优化实例应用表明：该算法有较快的收敛速度,能够较准确地寻得全局最优。     

混沌粒子群算法及其在生化过程动态优化中的应用

化工过程的动态优化,大多较为复杂,有相当的难度.新近发展的粒子群优化算法,基于群智能机理,适于求解连续问题,但它不具备遍历特性,影响了全局搜索能力.本文拟引入混沌机制,以混沌变量的遍历性改进粒子群算法,使其更全面地获取目标函数的有用信息,并反映到逐代更新的个体极值和群体极值中,可更有效地带领粒子群移向最优解,提高了全局搜优效率.由此构建为混沌粒子群算法,经多个性能测试,表明其搜索能力优于经典粒子群算法,引入混沌机制是有效的.将其用于Park-Ramirez生物反应器补料流率的动态优化,也取得了满意的效果.     

微粒群优化算法研究进展

围绕微粒群优化(PSO)算法的原理、特点、改进、应用等方面进行全面综述,介绍针对复杂环境的PSO研究内容,包括多目标、约束、离散和动态优化等,提出PSO有待进一步研究的若干方向和内容。     

一种基于PSO算法的闭环辨识方法

闭环模型辨识一直是工业先进控制领域中的一个主要课题。而现在基于粒子群优化算法PSO的辨识,大多都是连续开环系统的辨识。离散闭环模型辨识在计算机控制、运算量等方面比连续开环系统的辨识有较大的优势。文中讨论了PSO的时变惯性权重算法与参数初值的设置和选择方法。通过仿真实验表明,PSO与最小二乘递推算法相比,在有效性和一致性方面,有着明显的优势。PSO算法是一种有效地解决优化问题的群集智能算法,它的突出特点是算法中需要选择的参数少,程序实现简单,并在种群数量、寻优速度等方面较其他进化算法具有一定的优势。该方法在实际项目的应用中取得了较好的效果,应用前景广阔。     

Fatty Acid and Phytochemical Compositions of Plantago Seed Oils and Their Functionalities

To develop the potential applications of Plantago plants, seed oils were extracted from 14 cultivars of Plantago around China. Their fatty acid profiles, tocopherols, carotenoid compositions, anti‐inflammatory and antioxidant properties were also investigated. The Plantago seed oils (PSO) were abundant in linolenic acid from 11.12 to 29.36 g/100 g oil and had low fatty acid ratio of n‐6/n‐3 ratio matched with the dietary recommendations. The tocopherol contents of PSO ranged from 693.25 to 3708.80 μg/g and the lutein contents ranged from 2.29 to 26.68 μg/g. The PSO showed significant inhibitory effects on TNF‐α, IL‐1β, and COX‐2 mRNA expression in RAW 264.7 mouse macrophage cells induced by LPS. In addition, the properties on scavenging DPPH, oxygen and hydroxyl radicals indicated that PSO had potential antioxidant properties. The results could develop PSO as novel functional foods to improve human health.     

微粒群算法在干燥器多变量优化设计中的应用

建立了以具有废气循环的回转干燥系统年总费用为目标函数的优化设计数学模型,在此基础上探讨了惯性权因子对微粒群算法性能的影响,并应用微粒群算法求解干燥器优化设计数学模型,对干燥器出口废气温度与循环比进行优化设计。结果表明,带动态非线性惯性因子的微粒群算法对求解多变量的干燥优化设计问题具有方法简单、所需微粒群规模小、收敛速度快等特点;采用部分废气循环并进行优化设计对干燥系统的节能具有十分重要的意义,对湿空气出口温度和废气循环比进行优化设计,其年总费用比无废气循环的常规设计节省18．2%,比循环比为0．2时的常规设计节省12．6%。     

Analysis of the particle swarm algorithm in the optimization of a three-phase slurry catalytic reactor

The Particle Swarm Optimization (PSO) method was employed to optimize an industrial chemical process characterized by being difficult to be optimized by conventional deterministic methods. The chemical process is a three phase catalytic slurry reactor (tubular geometry) in which the reaction of the hydrogenation of o-cresol producing 2-methyl-cyclohexanol is carried out. The optimization problem was formulated considering as input variables the operating conditions of the reactor and as objective function the maximization of productivity, subject to the environmental constraint of conversion. The process was represented by a multivariable non-linear rigorous mathematical model and in order to solve the optimization problem, the performance of the PSO algorithm was evaluated considering four sets of parameters values suggested by the literature. PSO demonstrated to be efficient and robust to solve the constrained optimization problem, independently of the values of the PSO parameters. The solution of the rigorous mathematical model of the reactor was associated with a high computational burden, and although the PSO algorithm presented high rate of convergence, the attempt to make possible the optimization in a timeframe suitable to real time applications failed because the algorithm lost robustness (fraction of the number of runs the algorithm reached the optimization goal) when run with a reduced number of function evaluations. Therefore, if this type of application is desired, simplified mathematical models with fast and simple numerical methods must be preferred.     

Transparent luminescent bulk nanocomposites of polysiloxane embedded with CdS nanocrystallines by a direct dispersion process

Transparent luminescent bulk nanocomposites of polysiloxane (PSO) embedded with semiconductor nanocrystals (NCs) have been fabricated by the direct dispersion of CdS NCs in alkyl-(poly)siloxane (APS) followed by co-polymerization. The non-polar characteristics of the APS precursor are compatible with the CdS NC surface (oleylamine), which allows the direct dispersion of the CdS NCs without the need of any surfactant exchange. Chemical crosslinking of the NC-APS dispersion via hydrosilylation between Si-H and the vinyl group in APS immobilizes the CdS NCs in the polysiloxane network. Net-shaped three-dimensional bulk transparent polysiloxane/CdS NC composites were obtained by liquid casting of the NC-precursor dispersion and chemical crosslinking. The PSO/CdS NC composites show visible luminescence under ultraviolet excitation and the luminescent color is tunable from blue to red by controlling the NC concentration in the composite. Photoluminescence spectral analyses reveal the origin of the luminescence as being from the defect emission of the CdS NCs (550-900 nm) and an emission from the PSO matrix (380-550 nm). The luminescent spectra covered a wide range from the ultraviolet to the near-infrared region. The luminescence of the PSO/CdS NC nanocomposites was stable without any apparent degradation after exposure to air for a long time. This simple direct dispersion process is feasible for the fabrication of luminescent nanocomposites with useful optical properties for potential applications in optics and photoelectron devices.     

The synthesis of the particulates Al matrix composites by the compocasting method

Many scientific and engineering applications require optimization methods to find more than one solution to multimodal optimization problems. This paper presents an experimental and computational investigation on the influences of the secondary mechanical processing with different reductions on the porosity, microstructure and mechanical properties of compocast high strength and highly uniform Al matrix composites. Multiple solutions to the problems are located and refined using a new particle swarm optimization (PSO) technique. The technique, NichePSO, extends the inherent unimodal nature of the standard PSO approach by growing multiple swarms from an initial particle population. Each subswarm represents a different solution or niche; optimized individually. The outcome of the NichePSO algorithm is a set of particle swarms, each representing a unique solution. The as-cast composite exhibited higher porosity content as compared with the composite in the rolled conditions. Furthermore, the tensile strength of the composites increased considerably by increasing the reduction ratio in the cold rolling process.     

Microwave-absorption properties of SiOC ceramics derived from novel hyperbranched ferrocene-containing polysiloxane

In this contribution, we design a novel strategy to synthesize SiOC ceramics by pyrolysis of hyperbranched ferrocene-containing polysiloxane (HBPSO-VF) which are synthesized by the reaction of polysiloxane (PSO) with 1,1′-Bis(dimethylvinylsilyl)ferrocene (VF). This SiOC ceramics show much lower crystallization temperature because of the capability of HBPSO-VF to incorporate metallic iron into the backbone of PSO. The usage of HBPSO-VF offers enhanced ceramic yield of 83 wt% at 1200 °C due to the deep cross-linking of hydrosilylation. Nano-sized SiC and turbostratic carbons are separated from amorphous SiOC phase when it is annealed at 1100 °C, while crystallization temperature is 1400 °C when PSO is used as polymer precursors. The minimum reflection coefficient (RC_min) of this nanocrystal-containing ceramic reaches −46 dB, exhibiting a promising prospect as a kind of electromagnetic wave (EMW) absorbing materials. This method also can be further extended to develop other functional Si-based polymer derived ceramic (PDC) systems for EMW absorption and shielding applications.     

复合粒子群优化算法在模型参数估计中的应用

化工非线性模型的参数估计是较为困难的寻优问题，经典方法常会陷入局部极值。粒子群算法操作简便、容易实现且全局搜索功能较强，适用于非线性参数估计。但其参数值的确定与问题相关，若设定不当，会严重影响全局搜索的性能。今提出引入遗传算法，在粒子群算法的搜索过程中，逐代优选参数，包括惯性权值，加速常数，以此构建为复合粒子群优化算法。分析与测试表明，其全局搜索性能有显著改善。进一步的工作又将两种粒子群算法成功地应用于重油热解模型的参数估计。采用复合粒子群优化算法估计参数构建的重油热解模型，其预报相对误差比常规粒子群优化算法降低了8.97％，比简单遗传算法降低了23.21％，效果明显。     

变邻域宽度的爬山微粒群优化算法及其应用

微粒群优化算法（particle swarm optimization algorithm,PSO）是由Kennedy和Eberhart 1995年提出的进化计算算法．PSO简单且具有许多良好的优化性能,但对一些复杂优化问题存在容易陷入局部极值的缺陷．本文提出一种变邻域宽度的爬山微粒群优化算法（hill-climbing PSO with variable width neighborhood,vwnHCPSO）,并用5种测试函数进行测试和比较,然后将vwnHCPSO用于催化裂化装置（FCCU）主分馏塔轻柴油闪点软测量．