Particle swarm optimization ( pso ) is inspired by social behavior of bird flocking or fish schooling . it is a population - based , self - adaptive search optimization technique . pso is simple in concept , few in parameters , and easy in implementation 粒子群优化算法( particleswarmoptimization , pso算法)源于鸟群和鱼群群体运动行为的研究,是一种基于种群搜索策略的自适应随机算法,是进化计算领域中的一个新的分支。
The problem of information weak consistency must be considered by the load balancing algorithm whn it performs schedu1ing . in order to using the weak consistency information effectively , we propose a smallest k - subset random algoritlun ( skr and conduct a lot of simulation 为了有效利用系统中的弱一致性信息，使之成为负国防科学技术大学研究生院学位论文载平衡调度的决策依据，我们提出了一种基于信息弱一致性模型的最小k子集随机算法，并进行了性能模拟分析。
According to the fluctuating property of wind , on the stochastic vibration finite element theory , some calculation formulas of the response , including the along - wind response and the across - wind response , are obtained , the ideology of the torsion vibration in the double - tower structure and the ideology of the double - tower structure ' s dynamical wind load are also put forward at the same time . by using the stochastic vibration module of the general finite element program , the paper analyses the performance of the wind resistance more deeply and fully , discussing the influence on the double - tower structure from the diversification of the different component . the variance law of dynamic displacement and wind - induced vibration coefficient is derived through analyzing structure stochastic vibration , so is the effect of several parameters on dynamic displacement , acceleration and wind - induced vibration coefficient 从结构风振的基本理论出发，通过对脉动风功率谱的分析，推导了结构风振响应(顺、横风向)的计算公式，同时提出了双塔结构的扭转振动和动力风荷载计算的思想，在结合通用有限元程序的随机振动模块的情况下，对双塔结构的抗风性能作了深入、全面研究，探讨了双塔结构构件截面的变化对其抗风性能的影响，同时探讨了顺风向风速谱的选取、风速、参与计算的振型数、风与结构耦合作用是否考虑等因素对分析结果的影响，比较了规范风振系数的计算与随机算法结构的异同，并讨论了原因，得出一些有意义的结论。
Because our research is about optimized layout of entities whose quantity is arbitrary , it ’ s difficult to find the applicable meta - heuristics rules . and in random algorithms , simulated anneal and simple genetic algorithm are not only inefficient , but also can not guarantee the completeness of the results 因为本文研究的是任意多个实体的同时优化摆放问题,所以很难得到适用的启发式规则;而随机算法中的模拟退火、简单遗传算法等效率较低,并且不能保证算法的完备性。
Firstly , by selecting and defining the resource load indexes rationally , lbmr take into consideration the effect on the resource usage of source node and destination node by the migrant . by means of vector operations , lbmr can use multiply resources in harmony . secondly , lbmr adopts the smallest k - subset random algorithin , and the load index information is provided by the information cache Lbmr算法的主要思想有三点：第一，基于合理选择和定义的资源负载向量，综合考虑进程迁移对源节点和目标节点资源利用率的影响，通过向量运算协调多种资源的平衡使用；第二，基于最小k子集随机算法，采用信息cache提供的负载信息作为负载平衡算法位置策略的选择依据；第三，基于进程生命时间的概率分布，根据迁移进程减速( slowdown )数学期望值的改善程度，选择适合于迁移的进程。
