Environmental Influences on Fish Migration in a Hydropeaking River

Fish have evolved traits and life history characteristics that enable them to survive, exploit and depend on the flow regime of rivers, particularly the timing and predictability of flows for spawning and rearing their young. It is unclear to what degree pulsed flows from hydropower facilities and other environmental variables influence migratory behaviours. We used Dual Frequency Identification Sonar in the Michipicoten River, Canada, to address the relationship between fish migration and environmental factors with a focus on flow magnitude and fluctuation. In both 2007 and 2009, the peak of the Chinook (Oncorhynchus tshawytscha) and pink salmon (Oncorhynchus gorbuscha) migration occurred on 8 September. Mean water temperature on this date was 18 °C and precipitously dropped afterwards. The photoperiod was roughly 12‐h long with sunrise at 700 h and sunset at 2000 h. Most fishes moved upstream during the hours of darkness between 2000 and 600 h. The lowest counts of fish occurred from noon to just before sunset, whereas highest counts commonly occurred from 1 to 2 h after sunset. Fish moved upstream during all magnitudes of flow; however, there was a slight preference for larger flows in 2007 but not in 2009. Changes in flow magnitude occurred both during the day and night, with flows typically increasing during the day to meet electrical demand and decreasing at night. Most fishes moved upstream during periods of little to no change in flow. High flows and changing flows may deter salmon from moving up the Michipicoten River but not likely in a significant manner to cause energetic stress or harm. Other adverse effects of pulsed flows, however, must still be considered for spawning, hatching and rearing success. Copyright © 2014 John Wiley & Sons, Ltd.     

Importance of Reservoir Tributaries to Spawning of Migratory Fish in the Upper Paraná River

Regulation of rivers by dams transforms previously lotic reaches above the dam into lentic ones and limits or prevents longitudinal connectivity, which impairs access to suitable habitats for the reproduction of many migratory fish species. Frequently, unregulated tributaries can provide important habitat heterogeneity to a regulated river and may mitigate the influence of impoundments on the mainstem river. We evaluated the importance of tributaries to spawning of migratory fish species over three spawning seasons, by comparing several abiotic conditions and larval fish distributions in four rivers that are tributaries to an impounded reach of the Upper Paraná River, Brazil. Our study confirmed reproduction of at least 8 long‐distance migrators, likely nine, out of a total of 19 occurring in the Upper Paraná River. Total larval densities and percentage species composition differed among tributaries, but the differences were not consistent among spawning seasons and unexpectedly were not strongly related to annual differences in temperature and hydrology. We hypothesize that under present conditions, densities of larvae of migratory species may be better related to efficiency of fish passage facilities than to temperature and hydrology. Our study indicates that adult fish are finding suitable habitat for spawning in tributaries, fish eggs are developing into larvae, and larvae are finding suitable rearing space in lagoons adjacent to the tributaries. Our findings also suggest the need for establishment of protected areas in unregulated and lightly regulated tributaries to preserve essential spawning and nursery habitats. Copyright © 2014 John Wiley & Sons, Ltd.     

示踪试验在复杂岩溶区工程中的应用

以滇中引水工程某深埋长隧洞为依托,在较深入了解研究区水文地质及工程地质条件的基础上,选取线路对地下水环境影响较大的区域进行示踪试验,通过试验不仅较清晰地查清了地下水运移通道,而且较为准确地查明了研究区地下分水岭,最终为线路的选择提供了合理、准确、可行的依据。     

引入迁移学习的癫痫EEG信号自适应识别

在脑电图( EEG)信号识别中,EEG信号的采样环境、病人状态的多样性导致分类器训练所用的源域与分类器测试所用的目标域不匹配,分类器在目标域上表现不佳。为此,引入邻域适应策略,提出一种基于子空间相似度的改进主成分分析特征提取方法( SSM-PCA),在选择主成分时,考虑源域和目标域数据的几何和统计特性,并结合迁移学习分类器大间隔投射迁移支持向量机( LMPROJ),给出以SSM-PCA为基础的LMPROJ分类识别方法。实验结果表明,与结合PCA特征抽取技术和K近邻分类器实现的识别方法相比,该方法在识别正确率方面得到较大提升。     

基于时序模型迁移的间歇过程监测建模方法

针对工业生产过程频繁变更操作条件的需求, 提出一种基于时序模型迁移的方法, 通过从旧过程数据中挖掘并提取相似性信息, 可以快速建立新过程的监测模型. 将该建模方法与时序模型迁移相结合应用于青霉素发酵仿真过程, 实验结果验证了该方法的有效性.

     

基于人工免疫算法的增量式用户兴趣挖掘

了解用户兴趣是为用户提供个性化服务的关键.用户兴趣有短期兴趣和长期兴趣之分,且具有不稳定性.受人工免疫系统的启发,巧妙地将免疫应答过程应用于用户兴趣挖掘.首先将概率与时间相结合,提出“概念时序动态”的概念,以更好地刻画用户在一段时间内对同一兴趣的关注程度;然后基于人工免疫原理,建立抽取兴趣标签的分类器来提取用户兴趣标签;最后针对增量式学习,建立兴趣标签的“概念时序动态”,刻画出用户兴趣自首次出现以来受关注的程度,以此为依据来判断兴趣是否存在迁移及遗忘现象,并为每个兴趣标签附上权重.其主要贡献是创造性地将人工免疫原理应用于用户短期兴趣和长期兴趣的挖掘,并具有增量特性,可以很好地体现用户兴趣迁移特征,是一种自然完整的用户兴趣模型.实验结果表明,该学习模型能够很好地发现用户关注的领域,其平均精度和召回率分别达到79.5％和74.4％,是目前最贴近用户的兴趣挖掘模型.     

基于信息融合度传递的频域徙动入侵特征挖掘算法

在功率自激混合组合网络中,路由之间的相群特征相异性会产生谐振信号,因此需要有效挖掘入侵信号的频域徙动特征来实现对入侵信号的拦截。传统方法采用混合蛙跳算法挖掘入侵特征并且聚类中心矢量向模糊边缘贴近,因此搜索和挖掘精度不高。提出了一种基于混合蛙跳最优模因组信息融合度传递的频域徙动入侵特征挖掘算法。构建功率自激组合网络的系统模型和入侵信号数学模型,基于频域谐振慢变衰落幅度均衡原理,得到多源网络攻击源信号在相干点积功率累积尺度坐标,采用多普勒频移模糊搜索对入侵信号进行平滑处理,计算入侵信号的多普勒频移状态空间固有模态函数,得到入侵信号的频域特征包络幅度估计值。采用IIR滤波算法,对信号进行降噪滤波处理,提高信号的纯度,提出基于信息融合度传递的混合蛙跳入侵信号检测算法,优化特征挖掘结果,完成入侵信号的频域徙动特征挖掘算法改进。仿真实验结果表明,该算法能准确挖掘入侵信号的频域徙动特征,特征的波脊亮点明显,在低信噪比下提高了入侵信号的检测性能。     

基于Hadoop平台的用户行为挖掘

随着互联网发展带来的数据爆炸,使得 Web日志的数据量也越来越大,如何从海量的 Web 日志中挖掘有价值的信息成为了目前研究的热点。本文提出基于 Hadoop 集群框架对 Web 日志进行挖掘。实验结果表明,该集群系统既可以处理海量的 web 日志,同时也能够挖掘出有价值的信息,并证实了利用sqoop在 Hive仓库和传统数据库之间数据迁移的可行性。     

基于负载预测的虚拟机动态调度算法研究与实现

在云计算系统中为了实现负载均衡和资源的高效利用,需要在虚拟机粒度上对云计算系统进行调度,通过热迁移技术将虚拟机从高负载物理节点迁移到低负载物理节点。把负载预测技术和虚拟机动态调度技术相结合,提出了LFS算法,通过虚拟机历史负载数据对虚拟机未来的负载变化情况进行预测,然后根据预测结果对虚拟机进行调度,能够有效地避免云计算系统中高负载物理节点出现,实现负载均衡,提高资源使用率。     

Using task migration to improve non-contiguous processor allocation in NoC-based CMPs

In this paper, a processor allocation mechanism for NoC-based chip multiprocessors is presented. Processor allocation is a well-known problem in parallel computer systems and aims to allocate the processing nodes of a multiprocessor to different tasks of an input application at run time. The proposed mechanism targets optimizing the on-chip communication power/latency and relies on two procedures: processor allocation and task migration. Allocation is done by a fast heuristic algorithm to allocate the free processors to the tasks of an incoming application when a new application begins execution. The task-migration algorithm is activated when some application completes execution and frees up the allocated resources. Task migration uses the recently deallocated processors and tries to rearrange the current tasks in order to find a better mapping for them. The proposed method can also capture the dynamic traffic pattern of the network and perform task migration based on the current communication demands of the tasks. Consequently, task migration adapts the task mapping to the current network status. We adopt a non-contiguous processor allocation strategy in which the tasks of the input application are allowed to be mapped onto disjoint regions (groups of processors) of the network. We then use virtual point-to-point circuits, a state-of-the-art fast on-chip connection designed for network-on-chips, to virtually connect the disjoint regions and make the communication latency/power closer to the values offered by contiguous allocation schemes. The experimental results show considerable improvement over existing allocation mechanisms.