基于多尺度相关向量机的城市日用水量预测

为提高日用水量预测精度,提出一种基于多尺度相关向量机的预测模型。通过静态小波分解将用水量非平稳时间序列分解为不同尺度的平稳时间序列,然后在分解后的各子序列分别建立相关向量机回归模型进行预测,最后通过小波逆变换将各子序列预测结果整合得出原始用水量时间序列的预测值。在实例分析中分别利用多尺度关联向量机模型和单尺度相关向量机预测模型对实际用水量进行预测分析。结果表明,前者具有更高的预测精度,可应用于城市日用水量的预测。     

基于改进的神经网络与支持向量机的小流域日径流量预测研究

数据驱动水文模型可以在不考虑复杂物理过程的情况下,实现对数据种类较少的小流域日径流量的准确预测。本研究基于安徽省黄山市月潭水文监测站点2009-2012年的日径流量监测数据,分别构建粒子群寻优算法改进的神经网络(PSO-BPNN)以及支持向量机(PSO-SVM)模型。通过进行不同形式的模型结果比较发现,两类模型均有较好的拟合能力及泛化能力,其中基于三日流量数据的(PSO-SVM)模型具有最优模拟结果,可以考虑用于月潭流域日径流量的预测,实现流域内水资源的合理配置以及相关灾害的预防。     

对ZXDSL 9806H设备日常维护的探讨

随着IP业务、多媒体业务的飞速增长促使接入网进一步向宽带化和综合化的方向发展,于是出现了完全集成了宽带业务和窄带业务的复合型综合业务接入网,将传统电话业务与数据业务逐渐融合到下一代网络中,ZXDSL9806H设备是NGN网络环境中提供宽窄带一体化的接入网关,可以实现小容量的xDSL接入、窄带POTS接入,支持xPON接入,适应多种组网需求。随着光进铜退的快速发展,9806H在改建小区的应用越来越广泛,为了更好地维护和管理该设备,使用更加灵活,以及提高网络的可靠性,保证用户使用的效果,文章探讨了ZXDSL 9806H的设备原理及日常维护方法。     

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.     

上海市日供水量与气象要素的相关性分析及预测模型的建立

基于上海市2010-05-2013-12日供水资料和气象观测资料,利用数学统计方法分析日供水量与气象因子的关系,并建立日供水量预报模型。结果表明:一年之中,日供水量大体呈现"单峰单谷"的波动特征,夏季(7、8月份)日均供水最多,冬季(2月份)最少;日供水量具有显著的节假日效应;除降水外,日供水量与其它气象因子均有显著的相关性,且夏半年相关系数明显高于冬半年;日供水量与温度的关系最为密切,且与前一日最高温度的相关系数最高,同时,当温度大于20℃时,日供水量对于温度变化的响应极其敏感;供水预报模型精度较高,可为城市用水规划、合理调度的气象服务保障提供科学依据。     

基于支持向量机(SVM)的祁连山典型小流域日降水-径流模拟研究

及时准确的日径流预测在流域水资源的合理规划、利用及管理中具有十分重要的作用。本文以支持向量机(SVM)模型为基础,以祁连山典型小流域-排露沟流域为研究区域,建立了流域日降水-径流模型,对流域未来1~7 d的日径流量进行了模拟预测。为检验SVM模型的有效性,模拟结果与人工神经网络(ANN)模型预测结果进行了对比。结果表明:SVM和ANN均表现出了很高的精度;但相比于传统的ANN模型,SVM模型的预测精度显著提高。表明SVM模型在半干旱山区小流域径流预测中有更好的适用性,可以用于流域中长期日径流预测,是资料有限的条件下中长期日径流预测的有效工具。     

承担单一工业供水水库供水保证率推荐按月计算的方法

径流调节是确定河流治理开发方案及水利枢纽工程规模的重要环节,供水水库径流调节计算的任务是在来水确定的情况下,计算兴利库容、保证供水量和设计保证率三者之间的关系。根据《水利工程水利计算规范》（SL104-95）,对于综合利用的水库工程,当各用水部门设计保证率以不同方式表示时,先进行换算,统一以年保证率表示。但对承担单一供...     

Data-enabled learning and control algorithms for intelligent glucose management: The state of the art

External insulin administration is an effective way for patients with diabetes mellitus to regulate their blood glucose. Multiple daily injections (MDIs), sensor-augmented pump (SAP) and artificial pancreas (AP) are widely adopted approaches in insulin therapy. With the increasing popularity of continuous glucose monitoring (CGM) sensors, a large number of data-enabled learning and control algorithms have been developed for MDI, SAP and AP. In this paper, we perform a systemic review concerning the state-of-the-art methodologies that are developed for MDI, SAP and AP with feedback from CGM data or other available data, from a systems and control perspective. The review characterizes the traditional learning and control methods developed for the MDI, SAP and AP, including run-to-run control, proportional–integral–derivative control, fuzzy logic control and model predictive control, as well as the discussions on the roles of machine learning technologies in MDI, SAP and AP. Finally, potential future directions on the algorithm architecture design, a unified control framework for MDI, SAP and AP algorithm design and practical usage of the MDI, SAP and AP are discussed.     

Hyperprolactinemia in end‐stage renal disease and effects of frequent hemodialysis

Introduction: End‐stage renal disease is associated with elevations in circulating prolactin concentrations, but the association of prolactin concentrations with intermediate health outcomes and the effects of hemodialysis frequency on changes in serum prolactin have not been examined. Methods: The FHN Daily and Nocturnal Dialysis Trials compared the effects of conventional thrice weekly hemodialysis with in‐center daily hemodialysis (6 days/week) and nocturnal home hemodialysis (6 nights/week) over 12 months and obtained measures of health‐related quality of life, self‐reported physical function, mental health and cognition. Serum prolactin concentrations were measured at baseline and 12‐month follow‐up in 70% of the FHN Trial cohort to examine the associations among serum prolactin concentrations and physical, mental and cognitive function and the effects of hemodialysis frequency on serum prolactin. Findings: Among 177 Daily Trial and 60 Nocturnal Trial participants with baseline serum prolactin measurements, the median serum prolactin concentration was 65 ng/mL (25th–75th percentile 48–195 ng/mL) and 81% had serum prolactin concentrations >30 ng/mL. While serum prolactin was associated with sex (higher in women), we observed no association between baseline serum prolactin and age, dialysis vintage, and baseline measures of physical, mental and cognitive function. Furthermore, there was no significant effect of hemodialysis frequency on serum prolactin in either of the two trials. Discussion: Serum prolactin concentrations were elevated in the large majority of patients with ESRD, but were not associated with several measures of health status. Circulating prolactin levels also do not appear to decrease in response to more frequent hemodialysis over a one‐year period.     

Linux EXT4文件系统分析

介绍EXT4文件系统的原理和数据结构,重点与EXT3文件系统做了比较,通过分析WRITE操作来解释EXT4文件系统的工作,为用户选择Linux内核提供参考。