Hybrid Hierarchical Particle Swarm Optimization with Evolutionary Artificial Bee Colony Algorithm for Task Scheduling in Cloud Computing

Task scheduling plays a key role in effectively managing and allocating computing resources to meet various computing tasks in a cloud computing environment.Short execution time and low load imbalance may be the challenges for some algorithms in resource scheduling scenarios.In this work,the Hierarchical Particle Swarm Optimization-Evolutionary Artificial Bee Colony Algorithm(HPSO-EABC)has been proposed,which hybrids our presented Evolutionary Artificial Bee Colony(EABC),and Hierarchical Particle Swarm Optimization(HPSO)algorithm.The HPSO-EABC algorithm incorporates both the advantages of the HPSO and the EABC algorithm.Comprehensive testing including evaluations of algorithm convergence speed,resource execution time,load balancing,and operational costs has been done.The results indicate that the EABC algorithm exhibits greater parallelism compared to the Artificial Bee Colony algorithm.Compared with the Particle Swarm Optimization algorithm,the HPSO algorithmnot only improves the global search capability but also effectively mitigates getting stuck in local optima.As a result,the hybrid HPSO-EABC algorithm demonstrates significant improvements in terms of stability and convergence speed.Moreover,it exhibits enhanced resource scheduling performance in both homogeneous and heterogeneous environments,effectively reducing execution time and cost,which also is verified by the ablation experimental.

Hierarchical hesitant fuzzy K-means clustering algorithm

Due to the limitation and hesitation in one＇s knowledge, the membership degree of an element to a given set usually has a few different values, in which the conventional fuzzy sets are invalid. Hesitant fuzzy sets are a powerful tool to treat this case. The present paper focuses on investigating the clustering technique for hesitant fuzzy sets based on the K-means clustering algorithm which takes the results of hierarchical clustering as the initial clusters. Finally, two examples demonstrate the validity of our algorithm.

Dynamic load balancing based on restricted multicast tree in triplet-based hierarchical interconnection network

To solve the load balancing problem in a triplet-based hierarchical interconnection network（THIN） system, a dynamic load balancing （DLB）algorithm--THINDLBA, which adopts multicast tree （MT）technology to improve the efficiency of interchanging load information, is presented. To support the algorithm, a complete set of DLB messages and a schema of maintaining DLB information in each processing node are designed. The load migration request messages from the heavily loaded node （HLN）are spread along an MT whose root is the HLN. And the lightly loaded nodes（LLNs） covered by the MT are the candidate destinations of load migration; the load information interchanged between the LLNs and the HLN can be transmitted along the MT. So the HLN can migrate excess loads out as many as possible during a one time execution of the THINDLBA, and its load state can be improved as quickly as possible. To avoid wrongly transmitted or redundant DLB messages due to MT overlapping, the MT construction is restricted in the design of the THINDLBA. Through experiments, the effectiveness of four DLB algorithms are compared, and the results show that the THINDLBA can effectively decrease the time costs of THIN systems in dealing with large scale computeintensive tasks more than others.

Hierarchical Bayesian Calibration and On-line Updating Method for Influence Coefficient of Automatic Dynamic Balancing Machine

Measurement error of unbalance＇s vibration response plays a crucial role in calibration and on-line updating of influence coefficient（IC）. Focusing on the two problems that the moment estimator of data used in calibration process cannot fulfill the accuracy requirement under small sample and the disturbance of measurement error cannot be effectively suppressed in updating process, an IC calibration and on-line updating method based on hierarchical Bayesian method for automatic dynamic balancing machine was proposed. During calibration process, for the repeatedly-measured data obtained from experiments with different trial weights, according to the fact that measurement error of each sensor had the same statistical characteristics, the joint posterior distribution model for the true values of the vibration response under all trial weights and measurement error was established. During the updating process, information obtained from calibration was regarded as prior information, which was utilized to update the posterior distribution of IC combined with the real-time reference information to implement online updating. Moreover, Gibbs sampling method of Markov Chain Monte Carlo（MCMC） was adopted to obtain the maximum posterior estimation of parameters to be estimated. On the independent developed dynamic balancing testbed, prediction was carried out for multiple groups of data through the proposed method and the traditional method respectively, the result indicated that estimator of influence coefficient obtained through the proposed method had higher accuracy; the proposed updating method more effectively guaranteed the measurement accuracy during the whole producing process, and meantime more reasonably compromised between the sensitivity of IC change and suppression of randomness of vibration response.

A Multilevel Hierarchical Parallel Algorithm for Large-Scale Finite Element Modal Analysis

The strict and high-standard requirements for the safety and stability ofmajor engineering systems make it a tough challenge for large-scale finite element modal analysis.At the same time,realizing the systematic analysis of the entire large structure of these engineering systems is extremely meaningful in practice.This article proposes a multilevel hierarchical parallel algorithm for large-scale finite element modal analysis to reduce the parallel computational efficiency loss when using heterogeneous multicore distributed storage computers in solving large-scale finite element modal analysis.Based on two-level partitioning and four-transformation strategies,the proposed algorithm not only improves the memory access rate through the sparsely distributed storage of a large amount of data but also reduces the solution time by reducing the scale of the generalized characteristic equation(GCEs).Moreover,a multilevel hierarchical parallelization approach is introduced during the computational procedure to enable the separation of the communication of inter-nodes,intra-nodes,heterogeneous core groups(HCGs),and inside HCGs through mapping computing tasks to various hardware layers.This method can efficiently achieve load balancing at different layers and significantly improve the communication rate through hierarchical communication.Therefore,it can enhance the efficiency of parallel computing of large-scale finite element modal analysis by fully exploiting the architecture characteristics of heterogeneous multicore clusters.Finally,typical numerical experiments were used to validate the correctness and efficiency of the proposedmethod.Then a parallel modal analysis example of the cross-river tunnel with over ten million degrees of freedom(DOFs)was performed,and ten-thousand core processors were applied to verify the feasibility of the algorithm.

Toward a Multi-Hop, Multi-Path Fault-Tolerant and Load Balancing Hierarchical Routing Protocol for Wireless Sensor Network

This paper describes a novel energy-aware multi-hop cluster-based fault-tolerant load balancing hierarchical routing protocol for a self-organizing wireless sensor network (WSN), which takes into account the broadcast nature of radio. The main idea is using hierarchical fuzzy soft clusters enabling non-exclusive overlapping clusters, thus allowing partial multiple membership of a node to more than one cluster, whereby for each cluster the clusterhead (CH) takes in charge intra-cluster issues of aggregating the information from nodes members, and then collaborate and coordinate with its related overlapping area heads (OAHs), which are elected heuristically to ensure inter-clusters communication. This communication is implemented using an extended version of time-division multiple access (TDMA) allowing the allocation of several slots for a given node, and alternating the role of the clusterhead and its associated overlapping area heads. Each cluster head relays information to overlapping area heads which in turn each relays it to other associated cluster heads in related clusters, thus the information propagates gradually until it reaches the sink in a multi-hop fashion.

关于结合层次聚类和K-means算法进行聚类的研究

为了解决进行K-means聚类时类数的自动选择和Hierarchical聚类在处理大量高维数据时时间效率低的问题,在Kmeans聚类算法的基础上结合Hierarchical聚类算法,提出了一种基于集体智慧编程方法的用于处理大量数据时动态选取K值的聚类模型。实验结果表明该算法比K-means聚类具有更好的聚类效果,同时解决了Hierarchical聚类方法时间效率低的问题。本模型通过K-means聚类生成适量的类簇,再利用Hierarchical聚类对这些类再进行聚类,最后经过剪枝得到合适的聚类结果,以此实现动态选取K值。

Analyzing the Urban Hierarchical Structure Based on Multiple Indicators of Economy and Industry: An Econometric Study in China

For a city,analyzing its advantages,disadvantages and the level of economic development in a country is important,especially for the cities in China developing at flying speed.The corresponding literatures for the cities in China have not considered the indicators of economy and industry in detail.In this paper,based on multiple indicators of economy and industry,the urban hierarchical structure of 285 cities above the prefecture level in China is investigated.The indicators from the economy,industry,infrastructure,medical care,population,education,culture,and employment levels are selected to establish a new indicator system for analyzing urban hierarchical structure.The factor analysis method is used to investigate the relationship between the variables of selected indicators and obtain the score of each common factor and comprehensive scores and rankings for 285 cities above the prefecture level in China.According to the comprehensive scores,285 cities above the prefecture level are clustered into 15 levels by using K-means clustering algorithm.Then,the hierarchical structure system of the cities above the prefecture level in China is obtained and corresponding policy implications are proposed.The results and implications can not only be applied to the urban planning and development in China but also offer a reference on other developing countries.The methodologies used in this paper can also be applied to study the urban hierarchical structure in other countries.

流域初始水权分配的主从递阶双层优化方法

依托“流域-地区-行业”层次的初始水权分配模式,“流域-地区”层次影响制约“地区-行业”层次的初始水权分配,“流域-地区”和“地区-行业”层次的初始水权分配共同影响流域综合效益。现有研究主要聚焦“流域-地区”或“地区-行业”层次的初始水权分配,但将两个层次集成进行初始水权分配仍有待深化研究。为此,构建主从递阶双层优化方法,开展流域初始水权分配研究。具体可表述为:①针对“流域-地区”层次,建立流域综合效益优化函数与约束条件。构造适应性诊断准则和空间均衡性诊断准则,嵌入流域综合效益优化函数中,提高地区之间初始水权分配与经济社会发展的适应性、流域分水的空间均衡性。②针对“地区-行业”层次,建立地区综合效益优化的非线性目标规划模型,保障行业发展用水需求。并构造匹配性诊断准则嵌入非线性目标规划模型中,提高地区不同行业初始水权分配与行业发展的匹配性。③集成“流域-地区”和“地区-行业”层次,应用主从递阶双层优化方法,开展黄河流域初始水权分配实践探索。研究得出:通过黄河流域初始水权分配方案设计,得到黄河流域涉及省区水权分配比例与其经济社会发展综合指标加权值比例的比值,其中,内蒙古-陕西、陕西-宁夏、宁夏-山西、山西-河南、河南-甘肃、甘肃-青海、青海-山东分别为0.84、1.45、0.84、1.08、1.19、1.77、1.02,满足适应性诊断准则。黄河流域初始水权分配的基尼系数为0.10,属于“空间均衡度很高”,满足空间均衡性诊断准则。黄河流域涉及省区不同行业的水权配置与行业发展的匹配度均超过0.90,属于“高度匹配”。同时,黄河流域经济、社会、生态效益优化函数对应的满意度分别达到0.946、0.943、1.000,黄河流域涉及省区的经济、社会、生态效益优化函数对应的满意度均超过0.90。结果表明,该方法能够有效破解流域初始水权分配难题。为保障流域初始水权分配方案的有效落实,必须强化水资源刚性约束和双控行动,提升流域水利科技创新能力,推进流域产业用水结构合理化与产业结构高级化。

深度学习编译器模型训练负载均衡优化方法

对于计算密集型的人工智能(AI)训练应用,其计算图网络结构更加复杂,数据加载、计算图的任务划分以及任务调度的负载均衡性都会成为影响计算性能的关键因素。为了使深度学习编译器中模型训练应用的任务调度达到负载均衡的状态,提出了三种计算图负载均衡优化方法:第一,通过自动建立数据加载与模型训练的高效流水实现中央处理器和后端计算设备的负载均衡,提高了系统整体能效;第二,通过计算图的分层优化技术,实现计算图在后端设备执行调度时的负载均衡;最后,通过自动建立层间的高效流水提高后端设备的资源利用率。实验结果表明,计算图负载均衡优化方法实现了训练任务到底层硬件设备自动映射过程中系统的负载均衡,与Tensorflow、nGraph等传统的深度学习框架和编译器相比,在不同模型训练中通过任务调度负载均衡优化技术分别获得了2%~10%的性能提升,同时能够使系统整体的能耗降低10%以上。

基于分层架构的退役电池可重构均衡控制研究

利用退役电池构建储能系统,充分利用其剩余价值,是解决大量退役电池再利用的重要途径。然而,经过多次充放电影响,退役电池不一致性问题较新电池更为突出。在此背景下,首先,针对退役电池储能系统提出了分层式可重构均衡拓扑。其次,基于该拓扑提出了分层式均衡控制策略。考虑组内部和组间两个层次的均衡控制,组内以SOC和端电压为均衡变量,通过重构实现电池单体状态均衡,组间以SOC为均衡变量,通过重构保证全电池簇均衡。同时考虑多种运行工况:静置过程中通过自均衡提高电池组的均衡效率和可用容量,放电过程中通过拓扑结构变换维持输出电压稳定,充电过程中考虑能量损耗和充电速度合理优化充电电流,缩短充电时间,减少温升,延长电池使用寿命。最后,在Matlab中验证了所提拓扑与均衡策略在不同工况下的有效性。

无线传感器网络中一种基于聚合层次聚类的分簇路由算法

针对无线传感器网络中节点连接以及能量受限不足的问题,为了延长网络寿命,提出了一种基于AHC的分簇路由算法(HACCRA)。该算法首先运用AHC对网络节点分簇,接着为簇首选择、簇形成和路径构建分别定义了恰当的决策目标函数,运用能量阈值、提出距离阈值、并且路由过程优先考虑簇首节点之间的一对一连接,有效解决了路由算法中分簇和路由不衔接的问题。仿真结果表明,与JCR、ICR以及DCK-LEACH相比,HACCRA能够更好地实现网络节点的能耗均衡,保证网络数据传输的连接性,从而延长网络寿命。

基于平衡分层K均值的正交无监督大型图嵌入降维算法

为了降低大规模数据集降维的计算代价,提出一种基于平衡分层K均值的正交无监督图嵌入降维方法。该文给出局部保持投影和谱回归等价的充分必要条件;基于平衡分层K-means的锚生成策略,构建加快局部保持投影求解过程的特殊相似矩阵;再结合正交约束,提出正交化无监督大型图嵌入降维方法;在几种公开数据集上进行扩展实验,结果表明提出的方法能够对大规模数据集实现高效快速的降维。

能耗均衡与节能的自适应水下路由协议

针对水下传感器网络中节点能耗不均衡和能量有限的问题,提出一种能耗均衡与节能的自适应水下路由协议ECBES(energy consumption balanced and energy saving adaptive underwater routing protocol)。构建双区非均匀分层拓扑。基于能耗均衡因子,利用拓扑和节点剩余能量计算节点转发优先级,实现自适应转发节点选择,均衡网络能耗。与此同时,通过候选转发区域各分区域中节点参与转发数据包的比例确定次优候选转发区域,将次优候选转发区域作为初始策略,利用策略迭代思想确定最优候选转发区域,保证投递率的同时减少不同网络规模中重复数据包的转发,降低网络的整体能耗。仿真结果表明,ECBES相比VBF、ES-VBF和ALRP,在不同节点数量下,节点死亡率均最低,在保证数据包投递率的同时,能耗最少。

论我国公共数据开放的层次化规制

公共数据是政府、事业单位等部门在依法履行公共职能的过程中收集、处理形成的数据,具有体量大、价值高、种类多等特征。公共数据开放有助于激活数据价值,服务数字经济社会发展,是构建数字政府的重要环节。基于利益平衡理论,公共数据开放应当兼顾数据安全与公平利用,实现有区别的层次化规制。我国在公共数据开放治理上还存在数据分类标准不清、开放方式单一、权责失衡等局限,不利于公共数据开放目标的实现。以层次化开放为指导,我国应建立基于安全风险的数据分类分级制度、构建多样化的数据开放谱系规则、细化权责统一的公共数据开放义务体系,并通过数据开放平台的构建实现公私合作治理,促进公共数据的高效有序利用。

多模态深层次高置信度融合跟踪算法

为解决单目标跟踪中因目标外观及环境变化导致的跟踪失败问题,提出一种多模态深层次高置信度融合跟踪算法.首先构建目标颜色模型和基于双线性插值HOG特征形状模型的高维度多模态模型,之后对候选目标利用粒子滤波进行搜索.针对模型融合的难点,通过准确量化形状和颜色模型多种置信度并设计高置信度融合准则,以实现该多模态模型中不同置信度的深层次自适应加权平衡融合.最后针对模型更新参数固定的问题,设计非线性分级平衡更新策略.经过在OTB-2015数据集上的测试,发现该算法的平均CLE和OS在所有参照算法表现中均表现最佳,其值分别为30.57和0.609.此外,其FPS为15.67,满足了跟踪算法在一般情况下的实时性要求.在某些常见的特定场景中,其精确率、成功率指标在多数情况下的表现也超过了同类顶尖算法.

多约束混流线平衡与排序优化问题研究

针对G企业混流装配线设计不合理导致工位间负载不均衡及产品堆积现象,考虑工序与工位间匹配关系,节拍时间、工序优先关系等多约束,以减少工位数量、均衡工位间负载、减少产品等待时间为目标,建立混流装配线设计与产品排序问题的多目标混合整数规划模型,分别设计了分层算法和混合启发式算法;利用小规模算例验证了分层次算法的精确性,利用大规模算例验证了混合启发式算法高效性。基于本文算法对G企业装配线进行优化后,其工位间负载极差由15.6 min减少到0.1 min,优化后投产顺序对应的节拍时间降低了6 min。

一种大规模链式电池储能系统分层式并行均衡器

针对大规模链式电池储能系统,提出一种分层式并行均衡器。该均衡器采用两层主动均衡:第1层均衡可实现电池系统中所有单体电池同时进行能量均衡,均衡速度快且不受串联单体电池数目的影响;第2层均衡可实现来自不同电池组的多个电池单元的并行均衡充、放电,均衡速度快、效率高。该文对均衡器工作原理和均衡控制策略进行分析,为验证所提均衡方法的有效性,搭建由12个单体锂离子电池串联组成的电池系统实验平台进行均衡实验。理论和实验结果均表明,该均衡拓扑的均衡速度快、均衡效率高、模块化强、易扩展。

一种规模化混杂生产线缓冲区容量优化分配技术

针对传统技术难以解决规模化混杂生产线缓冲区容量分配问题(Buffer allocation problem, BAP),提出了一种规模化生产线递阶分解建模并行寻优技术(Hierarchical decomposition modeling parallel optimizing technique of large-scale production lines, HDMPOT).该技术结合混杂生产线系统综合方法与分解方法的技术思想,兼顾生产线平衡性与系统规模,将原系统递阶分解为包含虚拟生产线在内的n+1个子生产线系统,通过求解子系统的最优解构造原系统的渐近最优解,并在系统递阶建模阶段,提出了一种设备模糊聚类的辅助方式;同时,基于混杂生产线系统综合方法,提出了一种系统渐次综合的初解改进确定方法;并提出了一种通过构造动态步长来设计领域结构的改进型禁忌搜索算法(Simple tabu search, STS),对子系统进行并行寻优.最后,对技术算法的收敛性进行了证明.提出的生产线递阶分解建模并行寻优技术具有一般性,对受设备随机故障等随机事件影响的生产线,尤其是规模化生产线系统其他优化、控制问题也具有借鉴和参考价值.

广域分布式储能系统分层控制和SOC均衡控制研究

为了实现广域范围内分布式储能的统一灵活调控,本文研究基于储能云端控制器架构的分布式储能分层分区控制策略。中间层“分布式储能节点控制器”是核心环节,其接收上层云端调度信息实现对下层储能模块SOC的均衡控制,以解决储能电池过度充/过放电问题,从而提高蓄电池储能系统及电网运行可靠性。在传统均衡控制基础上,本文首先改进了一种荷电均衡控制方法,利用SOC差值修正中间层控制系数最可提高荷电均衡速度;为进一步提高荷电均衡速度和防止储能模块功率越限,本文提出了一种基于SOC差值的反正切函数均衡控制策略。最后,通过仿真验证了所提控制方法的正确性和可行性。