基于改进二进制人工蜂群的BP神经网络并行集成学习算法及其应用研究

BP神经网络算法具有寻优效率不高、易发生早熟且最终求解精度不够等特点,针对以上问题,文章提出一种基于改进二进制人工蜂群算法(Improved Binary Artificial Bee Colony Algorithm)的BP神经网络并行集成学习算法(IBABC-BP).首先,文章构建以高斯变异函数作为概率映射函数的离散二进制人工蜂群算法(IBABC),分析证明了算法的有效性,并通过在4个Benchmark标准测试函数上证明了其寻优精度和收敛速度较其他4种改进人工蜂群算法均有大幅提高;其次,将改进的二进制人工蜂群算法(IBABC)用于训练BP神经网络,设计了IBABC-BP并行集成学习算法;最后,将IBABC-BP算法用于雾霾评估预测,以合肥地区的雾霾历史数据作为仿真数据.实验结果表明,IBABC-BP算法在寻优精度和收敛速度上较原始BP算法、人工蜂群ABC-BP算法、遗传GA-BP算法等算法有明显的提升,可以有效地提高雾霾评估预测的准确性.     

改进GA机制融合的二进制蜂群算法研究

为了克服人工蜂群算法蜜源更新过程中的随机性并保留蜜源中个体序列合理的组合形式,通过分析基本蜂群算法更新公式的机理,提出一种改进GA(Genetic A1gorithm)机制融合的二进制蜂群算法.算法以二进制编码,首先依概率对任意两蜜源进行"去同存异"操作后随机排列,将排列结果放入到其中某个体中形成新个体.然后依概率进行二进制个体的"翻转"操作,上述两种操作从其本质上相当于GA的类交叉和类变异操作;其次利用GA机制收敛性的证明方式在理论上证明算法是收敛的.最后通过应用不同特性的多维基准函数和算法之间的比较验证改进蜂群算法具有良好的收敛能力和鲁棒性.     

全局较优解引导的人工蜂群算法

人工蜂群算法(ABC)是一种模拟蜜蜂群体寻找优良蜜源的群体智能优化算法.针对人工蜂群算法收敛速度较慢、探索能力较强而开发能力偏弱等问题,提出一种改进的蜂群算法.算法利用更多的较优蜜源位置的信息来引导采蜜蜂和跟随蜂的搜索行为.为了提高算法的全局收敛速度,通过基于混沌策略的方式生成初始化种群,并且在每一代侦察蜂阶段后对全部新蜜源进行反向再搜索.另外,每次蜜蜂邻域搜索之后,采用比较新旧蜜源的花蜜值(而非适应度值)的方法来更新蜜源位置.通过对14个标准测试函数进行仿真实验,结果表明所提出的算法能有效加快收敛速度,提高开发能力和解的精度.     

元胞人工蜂群算法及其在0-1规划问题中的应用

针对人工蜂群算法早熟收敛问题,基于元胞自动机原理和人工蜂群算法,提出一种元胞人工蜂群算法.该算法将元胞演化和人工蜂群搜索相结合,利用元胞及其邻居的演化提高了种群多样性,避免陷入局部最优解.经一系列典型0-1规划问题实例的仿真实验和与其他算法对比,验证了本算法的效果和效率,获得了满意的结果.     

求解多维0-1背包问题的人工鱼群算法

对于多维0-1背包问题,国内外学者提出了诸如模拟退火、遗传算法、蚁群算法以及其他启发式算法.给出一种新的智能寻优方法——人工鱼群算法.算法通过各人工鱼的局部寻优,从而在群体中体现出全局最优.描述了人工鱼群算法的具体步骤并编程实现,通过多维背包算例进行了求解测试,获得了满意的效果.     

模糊人工蜂群算法的多选择多维背包问题求解

针对传统人工蜂群算法早熟收敛问题,基于模糊化处理和蜂群寻优的特点,提出一种模糊人工蜂群算法。将模糊输入输出机制引入到算法中来保持蜜源访问概率的动态更新。根据算法计算过程中的不同阶段对蜜源访问概率有效调整,避免算法陷入局部极值。通过对多选择多维背包问题的仿真实验和与其他算法的比较,表明本算法可行有效,有良好的鲁棒性。     

基于人工蜂群算法的非线性系统模型参数估计

将人工蜂群算法用于非线性系统模型的参数估计,通过对谷氨酸菌体生长模型的参数估计进行验证,并与人工神经网络、遗传算法和微粒群算法的优化结果进行了比较.仿真试验结果表明:人工蜂群算法对非线性系统模型的参数估计精度高于人工神经网络、遗传算法和微粒群算法的参数估计精度,为非线性系统模型参数估计提供了一种有效的途径.     

动态空间调度的混合遗传算法

提出了一种基于混合遗传算法的动态空间调度方法。首先利用遗传算法产生多个可行的分段调度序列,再采用动态决定分段位置的启发式算法——平均最大空闲矩形策略对遗传算法产生的调度序列进行解码。同时以完工时间和平台利用率的加权和作为适应度函数,充分考虑了空间调度问题所特有的动态性和时空关联性。遗传进化过程收敛后得到近似最优解,实现了调度方案的全局优化。对船厂实际生产数据进行了实证分析以及与其它算法的对比分析,证明了所提方法在空间调度问题上的有效性和实用性。     

基于云自适应遗传算法的K-means聚类分析

现有的基于遗传算法的K-means聚类算法,利用遗传算法的全局优化性提高了K-means算法的寻优能力,收敛速度却过慢.为了解决上述问题,提出基于云自适应遗传算法的K-means聚类算法,利用云模型云滴的随机性和稳定趋向性设计遗传算法的交叉和变异概率,并在进化过程中引入K均值算子,以克服算法收敛速度过慢的问题.实验比较表明,算法具有较好的全局优化性,且收敛速度较快,提高了聚类算法解决物流管理中数据聚类工作的能力.     

0-1背包问题的蜂群优化算法

在项目决策与规划、资源分配、货物装载、预算控制等工作中,提出了0-1背包问题.0-1背包问题是组合优化中的典型NP难题,根据群集智能原理,给出一种基于蜂群寻优思想的新算法—蜂群算法,并针对0-1背包问题进行求解.经实验仿真并与蚁群算法计算结果作对比,验证了算法在0-1背包问题求解上的有效性和更快的收敛速度.     

A discrete artificial bee colony algorithm for the no-idle permutation flowshop scheduling problem with the total tardiness criterion

In this paper, we present a discrete artificial bee colony algorithm to solve the no-idle permutation flowshop scheduling problem with the total tardiness criterion. The no-idle permutation flowshop problem is a variant of the well-known permutation flowshop scheduling problem where idle time is not allowed on machines. In other words, the start time of processing the first job on a given machine must be delayed in order to satisfy the no-idle constraint. The paper presents the following contributions: First of all, a discrete artificial bee colony algorithm is presented to solve the problem on hand first time in the literature. Secondly, some novel methods of calculating the total tardiness from makespan are introduced for the no-idle permutation flowshop scheduling problem. Finally, the main contribution of the paper is due to the fact that a novel speed-up method for the insertion neighborhood is developed for the total tardiness criterion. The performance of the discrete artificial bee colony algorithm is evaluated against a traditional genetic algorithm. The computational results show its highly competitive performance when compared to the genetic algorithm. Ultimately, we provide the best known solutions for the total tardiness criterion with different due date tightness levels for the first time in the literature for the Taillard’s benchmark suit.     

基于人工蜂群算法的配送中心选址问题求解

采用人工蜂群算法对配送中心选址问题进行求解,给出食物源的编码方法,通过整数规范化,使算法能在整数空间内对问题进行求解.应用算法进行了仿真实验,并将结果与其它一些启发式算法进行了比较和分析.计算结果表明人工蜂群算法可以有效求解配送中心选址问题,同时也为算法求解其它一些组合优化问题提供了有益思路.     

多目标0-1规划问题的蜂群算法

针对多目标0-1规划问题,本文给出一种新型的智能优化算法——蜂群算法进行求解,并通过实例验证,与遗传算法、蚁群算法和元胞蚁群算法作了相应比较。就多目标0-1规划问题而言,蜂群算法能得到更多的Pareto解,说明了蜂群算法在解决该类问题上的有效性。     

蜂群算法求解支持模糊QoS约束的电子采购模型

基于Web的电子物流采购是电子商务的热点.通过将模糊量化的QoS约束加入到Web电子物流采购模型,构造一种带有QoS约束的Web电子物流采购模型,并设计一种改进的人工蜂群算法进行求解.模型基于现有Web采购系统的采购模式,融合了非功能性Web服务评价理论,着重将带QoS约束的电子物流采购选择过程与人工蜂群算法求解过程相结合,从而能快速准确地获得使整体利益最大的解.经仿真计算,验证了模型的有效性.     

An artificial bee colony algorithm for the capacitated vehicle routing problem

This paper introduces an artificial bee colony heuristic for solving the capacitated vehicle routing problem. The artificial bee colony heuristic is a swarm-based heuristic, which mimics the foraging behavior of a honey bee swarm. An enhanced version of the artificial bee colony heuristic is also proposed to improve the solution quality of the original version. The performance of the enhanced heuristic is evaluated on two sets of standard benchmark instances, and compared with the original artificial bee colony heuristic. The computational results show that the enhanced heuristic outperforms the original one, and can produce good solutions when compared with the existing heuristics. These results seem to indicate that the enhanced heuristic is an alternative to solve the capacitated vehicle routing problem.     

A DISCRETE AGE STRUCTURED SI MODEL OF TRACHEAL MITE (Acarapis woodi (RENNIE)) INFESTATION OF HONEY BEES (Apis mellifera L.) AT THE COLONY LEVEL

This paper presents a discrete age structured SI model of tracheal mite (Acarapis woodi (Rennie)) infestation of honey bees (Apis mellifera L.) at the colony level. The model describes the elevated honey bee attrition during winter and predicts that winter brood rearing should promote colony survival by decreasing the colony's infestation and increasing the colony's population. The model also predicts that swarming and the artificial removal of part of the colony's forager honey bees should reduce the colony's infestation level.     

A global best artificial bee colony algorithm for global optimization

The artificial bee colony (ABC) algorithm is a relatively new optimization technique which has been shown to be competitive to other population-based algorithms. However, there is still an insufficiency in the ABC algorithm regarding its solution search equation, which is good at exploration but poor at exploitation. Inspired by differential evolution (DE), we propose a modified ABC algorithm (denoted as ABC/best), which is based on that each bee searches only around the best solution of the previous iteration in order to improve the exploitation. In addition, to enhance the global convergence, when producing the initial population and scout bees, both chaotic systems and opposition-based learning method are employed. Experiments are conducted on a set of 26 benchmark functions. The results demonstrate good performance of ABC/best in solving complex numerical optimization problems when compared with two ABC based algorithms.