混合型粒子群优化算法研究

为了改进粒子群算法的性能,提出了融合其他算法优点的混合型粒子群算法。对三种主流的混合粒子群优化算法(基因粒子群、免疫粒子群、混沌粒子群)分别从混合目的、混合方式、实现步骤、算法优化性能等多个方面进行了研究,给出了这三种混合粒子群算法的优缺点及适用范围。     

离散微粒群优化算法的研究进展

首先,介绍了近年来出现的５种较为典型的离散ＰＳＯ,并分析了它们与基本ＰＳＯ 之间的联系和区别;然后,归纳了提高离散ＰＳＯ 优化性能的若干途径,并总结了离散ＰＳＯ 的应用现状;最后,探讨了离散ＰＳＯ 有待进一步研究的若干方向和内容．     

基于拓扑结构与粒子变异改进的粒子群优化算法

为使粒子群优化算法(PSO)优化过程的多样性与收敛性得到合理解决,以提高算法优化性能,基于种群拓扑结构与粒子变异提出两种粒子群改进算法RSMPSO和RVMPSO.改进算法将具有信息定向流动的闭环拓扑结构与星型拓扑结构或四边形拓扑结构相结合,促使粒子在前期寻优过程中具有较高的多样性,保证搜索的广度,而在后期满足粒子群的整体收敛性,保证寻优的精度.同时,将布谷鸟搜索算法(CS)中的偏好随机游走变异策略引入改进算法中,增强粒子跳出局部最优的能力.对标准测试函数的仿真实验表明,所改进的PSO算法与其他6个对比算法相比不仅操作简单,优化精度高,而且在算法收敛性及稳健性方面都有着更出色的表现.     

Hierarchical heterogeneous particle swarm optimization: algorithms and evaluations

Particle swarm optimization (PSO) has recently been extended in several directions. Heterogeneous PSO (HPSO) is one of such recent extensions, which implements behavioural heterogeneity of particles. In this paper, we propose a further extended version, Hierarchcial Heterogeenous PSO (HHPSO), in which heterogeneous behaviors of particles are enforced through interactions among hierarchically structured particles. Two algorithms have been developed and studied: multi-layer HHPSO (ml-HHPSO) and multi-group HHPSO (mg-HHPSO). In each HHPSO algorithm, stagnancy and overcrowding detection mechanisms were implemented to avoid premature convergence. The algorithm performance was measured on a set of benchmark functions and compared with performances of standard PSO (SPSO) and HPSO. The results demonstrated that both ml-HHPSO and mg-HHPSO performed well on all testing problems and significantly outperformed SPSO and HPSO in terms of solution accuracy, convergence speed and diversity maintenance. Further computational experiments revealed the optimal frequencies of stagnation and overcrowding detection for each HHPSO algorithm.     

A general exploratory projection pursuit network

Exploratory Projection Pursuit(EPP) is a statistical technique for finding interesting structure in high-dimensional data-sets. We introduce a negative feedback network which has been shown to perform EPP. We use the network with a novel but very simple activation function to search for different types of data structure where the form of the data structure is unspecified in advance.     

基于知识空间的分组式粒子群算法

为了提高粒子群算法中粒子搜索全局最优解的准确度,确保粒子的收敛性,提出了基于知识空间的分组式粒子群算法(KGPSO).该算法使用K-means算法对粒子群进行分组,利用较小的最大飞行速度(Vmax)加强粒子在组内的局部搜索能力,并将"知识空间"的概念带入到分组中,由知识空间中的粒子来引导群中粒子前往更好的解空间搜索.实验结果表明,KGPSO算法在测试函数的表现整体优于过去学者提出的标准PSO,HPSO、FPSO.     

Neutrino-like particle for particle swarm optimization

The real-world optimal problems frequently encountered by various industries are the nonlinear constrained optimization problems (NCOPs), where the constraints represent the limitations of practical resources. Many researchers have attempted to improve particle swarm optimization (PSO) in the past decades; however, in solving the NCOPs, the PSO-based approaches often cause premature convergences. The problem-specific constraints frequently generate many infeasible regions that block the movements of particles. The particles' behavior causes the exploration abilities of particles that tend to weaken along with time. The decreasing of exploration ability often comes from the particle becoming stagnant or moving unusefully. This study proposes a neutrino-like particle (NLP) with adaptive NLP hyperparameters that simulate the natural neutrino behavior. The proposed NLPs can be embedded in the PSO-based approaches for overcoming premature convergence. The experiment results demonstrate that all referenced PSO-based methods with the NLPs improved significantly compared with those without the NLPs to solve the NCOPs. All referenced PSO-based methods that embedded the NLPs also significantly outperform four recent strong algorithms in most IEEE CEC 2020 benchmark problems. Therefore, the proposed NLPs with adaptive NLP hyperparameters can effectively solve the premature convergences, reinforce the exploration ability, and maintain the exploitation capability for solving the NCOPs over the whole evolution process.     

Genetic particle swarm optimization for polygonal approximation of digital curves

Polygonal approximation is an important technique in image representation which directly impacts on the accuracy and efficacy of the subsequent image analysis tasks. This paper presents a new polygonal approximation approach based on particle swarm optimization (PSO). The original PSO is customized to continuous function value optimization. To facilitate the applicability of PSO to combinatorial optimization involving the problem in question, genetic reproduction mechanisms, namely crossover and mutation, are incorporated into PSO. We also propose a hybrid strategy embedding a segment-adjusting-and-merging optimizer into the genetic PSO evolutionary iterations to enhance its performance. The experimental results show that the proposed genetic PSO significantly improves the search efficacy of PSO for the polygonal approximation problem, and the hybrid strategy can accelerate the convergence speed but still with good-quality results. The performance of the proposed method is compared to existing approaches on both synthesized and real image curves. It is shown that the proposed hybrid genetic PSO outperforms the polygonal approximation approaches based on genetic algorithms and ant colony algorithms. The text was submitted by the author in English. Peng-Yeng Yin was born in 1966 and received his B.S., M.S. and Ph.D. degrees in Computer Science from National Chiao Tung University, Hsinchu, Taiwan, in 1989, 1991 and 1994, respectively. From 1993 to 1994, he was a visiting scholar at the Department of Electrical Engineering, University of Maryland, and the Department of Radiology, Georgetown University. In 2000, he was a visiting Associate Professor in the Visualization and Intelligent Systems Lab (VISLab) at the Department of Electrical Engineering, University of California, Riverside (UCR). He is currently a Professor at the Department of Information Management, National Chi Nan University, Nantou, Taiwan. His current research interests include image processing, pattern recognition, machine learning, computational biology, and evolutionary computation. He has published more than 70 articles in refereed journals and conferences. Dr. Yin received the Overseas Research Fellowship from the Ministry of Education in 1993 and Overseas Research Fellowship from the National Science Council in 2000. He is a member of the Phi Tau Phi Scholastic Honor Society and listed in Who’s Who in the World.     

基于遗传粒子群算法的高维复杂函数优化方法

针对高维复杂函数优化的特点,提出了一种遗传算法与粒子群算法相结合的主-从结构算法。算法中,主级为全局搜索的遗传算法;从级为局部邻域搜索的粒子群算法。通过主-从协调机制和从级转换函数设计,使算法不依赖复杂的编码方式和进化算子进行全局精确搜索。通过仿真和比较实验,验证了算法对高维复杂函数优化的有效性。     

Non-parametric particle swarm optimization for global optimization

In recent years, particle swarm optimization (PSO) has extensively applied in various optimization problems because of its simple structure. Although the PSO may find local optima or exhibit slow convergence speed when solving complex multimodal problems. Also, the algorithm requires setting several parameters, and tuning the parameters is a challenging for some optimization problems. To address these issues, an improved PSO scheme is proposed in this study. The algorithm, called non-parametric particle swarm optimization (NP-PSO) enhances the global exploration and the local exploitation in PSO without tuning any algorithmic parameter. NP-PSO combines local and global topologies with two quadratic interpolation operations to increase the search ability. Nineteen (19) unimodal and multimodal nonlinear benchmark functions are selected to compare the performance of NP-PSO with several well-known PSO algorithms. The experimental results showed that the proposed method considerably enhances the efficiency of PSO algorithm in terms of solution accuracy, convergence speed, global optimality, and algorithm reliability.     

A study on particle swarm optimization and artificial bee colony algorithms for multilevel thresholding

Segmentation is a critical task in image processing. Bi-level segmentation involves dividing the whole image into partitions based on a threshold value, whereas multilevel segmentation involves multiple threshold values. A successful segmentation assigns proper threshold values to optimise a criterion such as entropy or between-class variance. High computational cost and inefficiency of an exhaustive search for the optimal thresholds leads to the use of global search heuristics to set the optimal thresholds. An emerging area in global heuristics is swarm-intelligence, which models the collective behaviour of the organisms. In this paper, two successful swarm-intelligence-based global optimisation algorithms, particle swarm optimisation (PSO) and artificial bee colony (ABC), have been employed to find the optimal multilevel thresholds. Kapur's entropy, one of the maximum entropy techniques, and between-class variance have been investigated as fitness functions. Experiments have been performed on test images using various numbers of thresholds. The results were assessed using statistical tools and suggest that Otsu's technique, PSO and ABC show equal performance when the number of thresholds is two, while the ABC algorithm performs better than PSO and Otsu's technique when the number of thresholds is greater than two. Experiments based on Kapur's entropy indicate that the ABC algorithm can be efficiently used in multilevel thresholding. Moreover, segmentation methods are required to have a minimum running time in addition to high performance. Therefore, the CPU times of ABC and PSO have been investigated to check their validity in real-time. The CPU time results show that the algorithms are scalable and that the running times of the algorithms seem to grow at a linear rate as the problem size increases.     

双种群变异粒子群算法

利用变异机制可以增加遗传算法全局寻优能力的特性,结合惯性权值线性递减PSO算法具有较快收敛速度的优点,提出了一种双种群变异PSO算法,对该算法与其他PSO算法进行了比较,仿真结果表明其性能优越。     

Rock brittleness prediction through two optimization algorithms namely particle swarm optimization and imperialism competitive algorithm

Brittleness index (BI) is a significant rock parameter when dealing with projects performed in rocks. The main goal of this research work is to propose the novel practical models to predict the BI through particle swarm optimization (PSO) and imperialism competitive algorithm (ICA). For this aim, two forms of equations, i.e., linear and power are considered and the weights of these equations are optimized by PSO and ICA. In the other words, four predictive models, namely ICA linear, ICA power, PSO linear, and PSO power models are developed to predict BI in this study. In the modeling of the predictive models, 79 datasets are used, so that Schmidt hammer rebound number, wave velocity, density, and Point Load Index (Is₅₀) are selected as the independent (input) parameters and the BI values are considered as the dependent (output) parameter. Then, the performances of the proposed predicting models are checked using two error indices, namely coefficient correlation (R²) and root mean squared error (RMSE). The results showed that the PSO power model has superior fitting specification for the prediction of the BI compared to the other prediction models and is quite practical for use. As a result, linear and power models of PSO received higher performance prediction compared to ICA. PSO power (with R² train = 0.937, R² test = 0.959, RMSE train = 0.377 and RMES test = 0.289) showed the most powerful technique to predict BI of the granite samples.

     

改进的粒子群算法

为改善基本粒子群算法的搜索性能,针对粒子群算法随机性较强、收敛较慢的问题,利用数学中的外推技巧给出了两个新的粒子位置更新公式,由此构造出一种新的算法--强引导型粒子群算法.新算法对粒子位置更新加以引导,试图减少算法的随机性以提高搜索效率.用4个基准函数对新算法进行试验,结果表明,新算法在稳定性和收敛性上优于基本粒子群算法.     

Dispersed particle swarm optimization

In particle swarm optimization (PSO) literatures, the published social coefficient settings are all centralized control manner aiming to increase the search density around the swarm memory. However, few concerns the useful information inside the particles' memories. Thus, to improve the convergence speed, we propose a new setting about social coefficient by introducing an explicit selection pressure, in which each particle decides its search direction toward the personal memory or swarm memory. Due to different adaptation, this setting adopts a dispersed manner associated with its adaptive ability. Furthermore, a mutation strategy is designed to avoid premature convergence. Simulation results show the proposed strategy is effective and efficient.     

基于参考点的高维多目标粒子群算法

高维多目标优化问题一般指目标个数为4个 或以上时的多目标优化问题.由于种群中非支配解数量随着目标数量的增加而急剧增多,导致进化算法的进化压力严重降低,求解效率低.针对该问题,提出一种基于粒子群的高维多目标问题求解方法,在目标空间中引入一系列的参考点,根据参考点筛选出能兼顾多样性和收敛性的非支配解作为粒子的全局最优,以增大选择压力.同时,提出了基于参考点的外部档案维护策略,以保持最后所得解集的多样性.在标准测试函数DTLZ2上的仿真结果表明,所提方法在求解高维多目标问题时能够得到收敛性和分布性都较好的解集.     

Genetic based discrete particle swarm optimization for Elderly Day Care Center timetabling

The timetabling problem of local Elderly Day Care Centers (EDCCs) is formulated into a weighted maximum constraint satisfaction problem (Max-CSP) in this study. The EDCC timetabling problem is a multi-dimensional assignment problem, where users (elderly) are required to perform activities that require different venues and timeslots, depending on operational constraints. These constraints are categorized into two: hard constraints, which must be fulfilled strictly, and soft constraints, which may be violated but with a penalty. Numerous methods have been successfully applied to the weighted Max-CSP; these methods include exact algorithms based on branch and bound techniques, and approximation methods based on repair heuristics, such as the min-conflict heuristic. This study aims to explore the potential of evolutionary algorithms by proposing a genetic-based discrete particle swarm optimization (GDPSO) to solve the EDCC timetabling problem. The proposed method is compared with the min-conflict random-walk algorithm (MCRW), Tabu search (TS), standard particle swarm optimization (SPSO), and a guided genetic algorithm (GGA). Computational evidence shows that GDPSO significantly outperforms the other algorithms in terms of solution quality and efficiency.     

Example-based learning particle swarm optimization for continuous optimization

Particle swarm optimization (PSO) is a heuristic optimization technique based on swarm intelligence that is inspired by the behavior of bird flocking. The canonical PSO has the disadvantage of premature convergence. Several improved PSO versions do well in keeping the diversity of the particles during the searching process, but at the expense of rapid convergence. This paper proposes an example-based learning PSO (ELPSO) to overcome these shortcomings by keeping a balance between swarm diversity and convergence speed. Inspired by a social phenomenon that multiple good examples can guide a crowd towards making progress, ELPSO uses an example set of multiple global best particles to update the positions of the particles. In this study, the particles of the example set were selected from the best particles and updated by the better particles in the first-in-first-out order in each iteration. The particles in the example set are different, and are usually of high quality in terms of the target optimization function. ELPSO has better diversity and convergence speed than single-gbest and non-gbest PSO algorithms, which is proved by mathematical and numerical results. Finally, computational experiments on benchmark problems show that ELPSO outperforms all of the tested PSO algorithms in terms of both solution quality and convergence time.     

求解约束优化的改进粒子群算法

针对种群初始化时粒子过于集中和基本粒子群算法搜索精度不高的缺陷,提出了一种求解约束优化问题的改进粒子群算法。该算法引入佳点集技术来优化种群的初始粒子,使种群粒子初始化时分布均匀,因而种群具有多样性,不会陷入局部极值;同时使用协同进化技术使双种群之间保持通信,从而提高算法的搜索精度。仿真实验结果表明：将该算法用于5个基准测试函数,该算法均获得了理论最优解,其中有4个函数的测试方差为0。该算法提高了计算精度且鲁棒性强,可以广泛应用于其他约束优化问题中。     

改进的量子粒子群多目标优化算法

针对粒子群优化算法容易陷入局部极值点的问题,提出了一种新的量子比特粒子群算法,该算法采用Pareto支配关系来更新粒子的个体最优值和局部最优值;定义极大极小距离,并采用该距离方法裁减非支配解.实验结果表明该算法能更好地接近Pareto前沿且具有更好的分布性,更适合于求解复杂高维优化问题,是一种非常有潜力的多目标优化方法.