基于自适应排斥因子的改进粒子群算法

基本粒子群算法在求解复杂的多峰问题时,由于存在较多的局部最优解,算法极易出现早熟现象。为克服这一缺陷,采用蒙特卡洛(Monte Carlo)方法模拟了种群飞行轨迹,得出种群极易陷入局部最优解的原因;在此基础上,通过定义粒子间距离、粒子间最大距离和粒子间平均距离,提出一种自适应控制粒子自身最优位置和种群最优位置间距离的排斥因子(ARF),来提升种群跳出局部最优的能力。为测试提出策略的有效性,在60次独立运行时,基于ARF的改进PSO算法(ARFPSO)在Rosenbrock,Ackley和Griewank函数上所获得的最好值分别为53.82,2.1203和5.32E-004,都优于其他两种对比算法,这表明ARFPSO能有效地跳出局部最优解;算法的复杂度分析表明引入的策略没有增加计算复杂度。     

基于多种群粒子群算法和布谷鸟搜索的联合寻优算法

为了提高动态多种群粒子群(DMS-PSO) 算法的全局搜索能力, 将布谷鸟搜索算法(CS) 引入DMS-PSO 算法中, 提出DMS-PSO-CS 算法. 采用中位数聚类算法将整个种群动态划分为若干小种群, 各个小种群作为底层种群通过PSO 算法进行寻优, 再将每个小种群中的最优粒子作为高层种群的粒子通过CS 算法进行深度优化. 将所提出算法应用于CEC 2014 测试函数, 并与CS 算法和其他改进的PSO 算法进行比较. 实验结果表明, 所提出算法能够显著提高全局搜索能力和算法效率.

     

教与学信息交互粒子群优化算法

针对单一种群在解决高维问题中收敛速度较慢和多样性缺失的问题,提出了一种教与学信息交互粒子群优化(PSO)算法.根据进化过程将种群动态地划分为两个子种群,分别采用粒子群优化算法和教与学优化算法,同时粒子利用学习者阶段进行子种群之间信息交互,并通过评价收敛性和多样性指标让粒子的收敛能力和多样性在进化过程中得到平衡.与粒子群...     

融合多种策略的改进粒子群算法

为有效解决粒子群优化算法(Particle Swarm Optimization, PSO)容易陷入局部极值及进化后期收敛速度慢、精度低等缺点, 提出了一种融合多种策略的改进粒子群算法(Improved Particle Swarm Optimization, IPSO). 该算法包括以下4点改进:(1)采取分组控制策略, 按适应度值将种群分为优解组和劣解组, 优解组进行遗传交叉操作, 劣解组进行变异操作; (2)精英策略用来更新种群, 根据适应度值从经过交叉和变异操作后的种群及初始种群中选出前一半粒子作为新种群; (3)改进粒子学习模式, 充分利用种群信息, 以优良种群的均值代替个体最优位置;(4)引入概率控制来控制算法进入交叉和变异操作的概率. 测试函数的仿真结果表明, 与标准PSO及其改进算法相比, IPSO算法能有效兼顾全局探索和局部挖掘能力, 具有收敛速度快、求解精度高、避开局部最优解的优点.     

基于EM算法和PSO算法的混合优化算法

针对粒子群优化算法容易陷入局部最优解并且存在过早收敛的问题,将类电磁机制算法中的吸引-排斥机制引入到粒子群优化算法中,提出一种类电磁机制算法和粒子群优化算法的混合优化算法(EMPSO).首先按照基本粒子群优化算法的寻优方式对各粒子进行更新,再利用类电磁机制中的吸引-排斥机制对个体最优粒子和群体最优粒子进行移动,最后通过几个标准测试函数进行了测试,并与标准粒子群算法(PSO)、免疫粒子群算法(IPSO)、混沌粒子群算法(CPSO)进行对比.测试结果表明,改进算法提高了全局搜索能力和熟练速度,改善了优化性能.     

粒子置换的双种群综合学习PSO算法

针对粒子群算法(PSO)种群多样性低和易于陷入局部最优等问题,提出一种粒子置换的双种群综合学习PSO算法(PP-CLPSO).根据PSO算法的收敛特性和Logistic映射的混沌思想,设计并行进化的PSO种群和混沌化种群,结合粒子编号机制,形成双种群系统中粒子的同号结构和同位结构,其中粒子的惯性权重根据适应度值自适应调...     

改进的多目标粒子群优化方法

为了提高多目标粒子群算法(MOPSO)在Pareto前沿的收敛性和分布性,对传统MOPSO方法进行了改进.首先采用基于Pareto支配概念的适应值比例方法选择gbest,其次利用动态拥挤距离更新外部精英集,并通过对精英种群执行遗传操作,最后在粒子种群引入自适应的淘汰机制,加强粒子种群和精美种群的进化.典型测试函数的计算结果表明,该算法在收敛精度和分布性方面得到明显改善.     

基于搜索空间可调的自适应粒子群优化算法与仿真

针对收缩因子粒子群优化(CPSO)算法易陷入局部最优和发生过早收敛的问题.提出了基于搜索空间可调的自适应粒子群优化(APSO)算法.该算法根据种群早熟收敛程度和个体适应值,在CPSO算法停滞时,将全部粒子有效地划分在3类不同的搜索空间,使种群始终保持搜索空间的多样性,易于跳出局部最优,从而有效地改善了CPSO算法后期的寻优能力.     

基于食物链机制的动态多物种粒子群算法

针对粒子群优化(PSO)算法在解决多峰函数时容易陷入局部最优的问题,提出了一种基于食物链机制的动态多物种粒子群(DSPSO)算法。受生物界的启发,引入食物链机制来保证种群的多样性,并结合繁殖机制使得算法具有良好的优化性能。食物链机制中,整个标榜群被分为几个子种群,每个子种群都能够捕食另外一个子种群。通过一定概率发生的捕食现象使得标榜群得以进化,剔除对种群贡献小的粒子,并通过繁殖策略生成新的粒子。种群通过不断地进化保证了种群的多样性,同时通过剔除较差粒子的误导作用使算法的进化更有效率。为了验证算法的有效性,选择了包括偏移函数、旋转函数在内的10个测试函数来测试DSPSO算法的性能。实验结果表明DSPSO算法有着良好的寻优性能。与PSO、局部版本的粒子群(LPSO)算法、动态多群粒子群(DMS-PSO)算法和全面学习粒子群(CLPSO)算法相比,DSPSO算法不仅能够得到较高精度的解,而且还具有较高的可信度。     

均匀粒子群算法

由于粒子群算法(PSO)本质上的随机性,其搜索质量和速度也呈随机性.这使得普通的粒子群算法难以满足某些需要快速优化的工程需要.利用均匀设计方法产生PSO算法的初始种群（或关键代次种群）,可以使种群中的粒子在搜索空间分布更均匀,更好地保持分散性.算法中给出了4种种群的生成方案,通过测试和对比分析表明:基于值域分割的均匀设计种群生成法能使算法的搜索效果最好;算法可以在不丧失搜索精度和效率的前提下,提高搜索效率和搜索精度的稳定性,有效减少粒子聚集和早熟的发生.     

Prey predator hyperheuristic

Prey predator algorithm is a population based metaheuristic algorithm inspired by the interaction between a predator and its prey. In the algorithm, a solution with a better performance is called best prey and focuses totally on exploitation whereas the solution with least performance is called predator and focuses totally on exploration. The remaining solutions are called ordinary prey and either exploit promising regions by following better performing solutions or explore the solution space by randomly running away from the predator. Recently, it has been shown that by increasing the number of best prey or predator, it is possible to adjust the degree of exploitation and exploration. Even though, this tuning has the advantage of easily controlling these search behaviors, it is not an easy task. As any other metaheuristic algorithm, the performance of prey predator algorithm depends on the proper degree of exploration and exploitation of the decision space. In this paper, the concept of hyperheuristic is employed to balance the degree of exploration and exploitation of the algorithm. So that it learns and decides the best search behavior for the problem at hand in iterations. The ratio of the number of the best prey and the predators are used as low level heuristics. From the simulation results the balancing of the degree of exploration and exploitation by using hyperheuristic mechanism indeed improves the performance of the algorithm. Comparison with other algorithms shows the effectiveness of the proposed approach.     

融合混沌对立和分组学习的海洋捕食者算法

针对海洋捕食者算法存在收敛速度慢、不易逃出局部最优的缺点,提出了一种改进海洋捕食者算法。将混沌映射与对立学习策略相结合,在保证遍历性和随机性的同时,生成高质量的初始猎物种群。引入自适应t分布变异算子更新种群,增加种群多样性,避免陷入局部最优。对更新后的种群,按照适应度分为精英组和学习组,学习组向精英组猎物的平均维度进行学习,精英组内的猎物相互维度学习,进一步提高种群质量和搜索精度。选取15个测试函数,通过对比测试,验证了改进后的算法可以有效提高原算法的收敛速度和寻优精度。将改进后的算法应用于无线传感器网络覆盖优化,实验结果显示,改进后的算法提高了网络覆盖率,优化后的节点分布更加均匀。     

处理动态优化问题的捕食元胞遗传算法

根据自然界中的捕食关系,提出一种捕食策略来代替元胞遗传算法中的演化规则,并构建了基于捕食策略的元胞遗传算法以处理动态环境下的优化问题.在元胞空间中,捕食者对其捕食范围内的被捕者进行猎取并捕获其中最弱的一个.对捕食策略中种群规模的相互关系进行了研究,通过引入正交交叉算子进一步提高了算法的搜索能力.选择不同强度、复杂度的动态优化问题进行算法性能验证,所得结果表明新算法具有良好的处理动态优化问题的能力.     

Global qualitative analysis of a predator–prey system with Allee effect on the prey species

In this paper, the Allee effect is incorporated into a predator–prey model with linear functional response. Compared with the predator–prey model without the Allee effect, it is found that the Allee effect of the prey species increases the extinction risk of both the prey and predator. If the Allee effect of the prey species is strong and the mortality of the predator species is relatively low, then the prey and predator cannot coexist after the predator invasion. Moreover, it is shown that the model with Allee effect undergoes the heteroclinic loop bifurcation and subcritical and supercritical Hopf bifurcations. With the brokenness of the heteroclinic loop, a stable or unstable limit cycle will appear. The Allee effect of the prey species can lead to unstable or stable periodic fluctuations. It is also found that the positive equilibrium of the model could change from stable to unstable, and then disappear when the strength of Allee effect increases continuously from zero.     

多时滞SI模型的稳定性和Hopf分支

建立并分析了一个带有多时滞的捕食者和食饵都染病的SI模型,用特征根的方法求得了它的非负平衡点。通过分析得到当易感食饵的感染率小于某一阈值时,染病食饵和捕食者将最终灭绝。当易感食饵的感染率大于某一阈值且易感捕食者的转化系数小于某一阈值时,捕食者将最终灭绝。边界平衡点是局部渐近稳定的,随着食饵时滞的增加该平衡点由稳定变为不稳定,系统在该平衡点附近发生Hopf分支。捕食者的时滞对该平衡点的稳定性不产生影响。     

Ecoepidemics with infected prey in herd defence: the harmless and toxic cases

We consider a predator–prey population model with prey gathering together for defence purposes. A transmissible unrecoverable disease affects the prey. We characterize the system behaviour, establishing that ultimately either only the susceptible prey survive, or the disease becomes endemic, but the predators are wiped out. Another alternative is that the disease is eradicated, with sound prey and predators thriving at an equilibrium or through persistent population oscillations. Finally, the populations can thrive together, with the endemic disease. The only impossible alternative in these circumstances is predators thriving just with infected prey. But this follows from the model assumptions, in that infected prey are too weak to sustain themselves. A mathematical peculiarity of the model is the singularity-free reformulation, which leads to three entirely new dependent variables to describe the system. The model is then extended to encompass the situation in which ingestion of diseased prey is fatal for the predators and to the cases where the predators find the infected prey less palatable.     

带自控能力的捕食模型正解存在性与数值模拟

在齐次Dirichlet边界条件下,研究一类低密度食饵下,捕食者具有自控能力的捕食模型平衡态正解存在性。通过连续延拓意义下建立的连续算子,利用度理论给出了平衡态正解存在的充分条件,并对理论结果进行数值模拟。研究结果表明,只要捕食者和食饵的生长率适当大,则捕食者和食饵可以共存。     

Stationary patterns of a prey–predator system with a protection zone and cross-diffusion of fractional type

This paper is concerned with the stationary patterns of a prey–predator model with a protection zone and fractional type cross-diffusion for the prey. It is shown that the fractional type cross-diffusion has negative effects on the survival of the prey when the intrinsic growth rate of the predator is positive. Moreover, our mathematical analysis shows that, compared with the results obtained in K. Oeda (2011) and K. Oeda (2012), the large cross-diffusion coefficient and large growth rate of the predator species have some essentially different effects on the profiles of the solutions.     

ALife approach for body-behavior predator–prey coevolution: body first or behavior first?

We present the results of morphology–behavior predator–prey coevolution in a 3D physically simulated environment. The morphology and behaviors of virtual creature predators and prey are evolved using a genetic algorithm and random one-on-one encounters in a shared environment. We analyze the evolutionary dynamics on the basis of quantitative characterization of morphology and behavior. Specifically, we pose and answer the question: which precede the other, morphology or behavior, during the evolutionary acquisition of predator and prey strategies?     

Comparative Study of Prey Predator Algorithm and Firefly Algorithm

Metaheuristic algorithms are found to be promising for difficult and high dimensional problems. Most of these algorithms are inspired by different natural phenomena. Currently, there are hundreds of these metaheuristic algorithms introduced and used. The introduction of new algorithm has been one of the issues researchers focused in the past fifteen years. However, there is a critic that some of the new algorithms are not in fact new in terms of their search behavior. Hence, a comparative study in between existing algorithms to highlight their differences and similarity needs to be studied. Apart from knowing the similarity and difference in search mechanisms of these algorithms it will also help to set criteria on when to use these algorithms. In this paper a comparative study of prey predator algorithm and firefly algorithm will be discussed. The discussion will also be supported by simulation results on selected twenty benchmark problems with different properties. A statistical analysis called Mann—Whitney U 2 test is used to compare the algorithms. The theoretical as well as simulation results support that prey predator algorithm is a more generalized search algorithm, whereas firefly algorithm falls as a special case of prey predator algorithm by fixing some of the parameters of prey predator algorithm to certain values.