基于BP神经网络的金沟河灌区旬来水量预报

考虑到金沟河灌区径流量长期变化的因果关系较为复杂的特性,若采用常规中长期水文预报模型很难达到预测精度要求。结合实际水量数据,采用BP神经网络预测模型对金沟河灌区旬来水量进行预报,并进行模型预报精度验证。结果证明,BP神经网络预测模型完全可以用于中长期水文预报。     

基于BP神经网络的来水量预测模型

针对径流量长期变化的因果关系复杂特性,常规的中长期水文预报模型又很难满足精度要求,提出了基于BP神经网络的来水量预测模型。结合实际径流数据,验证了模型的预报精度,可用来进行中长期水文预报。     

相空间神经网络耦合模型在水文预测中的应用

混沌相空间理论和神经网络用于径流系统的中长期预测,较传统途径可以更多地利用时间序列中包含的丰富信息,更好地揭示水文动力学系统的规律.针对混沌时间序列,结合混沌分析理论和BP神经网络,建立了相空间重构和BP神经网络耦合预测模型.经实例研究初步表明,用神经网络拟合相空间相点演化的非线性关系是可行的;相空间神经网络耦合预测模型在水文中长期预报中的应用是可行的、合理的,有较好的预报精度和应用价值.     

基于小波去噪和FA-SVM的中长期径流预报

为优化支持向量机(Support Vector Machine,SVM)模型参数,提高中长期径流预报精度,建立了基于FA-SVM的中长期径流预报模型。该模型以样本中训练期均方差最小为目标函数,利用萤火虫算法(Firefly Algorithm,FA)对支持向量机主要参数(惩罚系数C、核函数参数g和不敏感损失系数ε)进行了优化。以岷江上游的紫坪铺水库为例,运用小波去噪法对各月径流序列进行数据预处理后,利用FA-SVM模型与BP神经网络模型进行了中长期径流预报。结果表明:(1)运用小波阈值法能够较好地滤除各月径流序列的系统噪声和测量噪声;(2) FA-SVM模型中长期径流预测效果较好,预报精度等级均在丙级以上;(3) FA-SVM模型的预报效果优于BP神经网络模型。     

花授粉算法-BP神经网络模型及其在月径流预报中的应用

通过5个典型高维函数对一种新型群智能优化算法——花授粉算法(FPA)进行仿真验证,并与粒子群优化(PSO)算法进行对比分析。针对BP神经网络初始权值、阈值参数难以确定的不足,利用FPA算法优化BP神经网络初始权值和阈值,提出FPA-BP预报模型,以清水江水文站12月月平均流量预报为例进行实例研究。结果表明:1FPA算法收敛精度优于PSO算法,具有较好的收敛速度和全局寻优能力;2FPA-BP模型对实例2001—2005年月平均流量预报的平均相对误差绝对值为2.38%(5次平均),预报精度优于PSO-BP、BP模型。利用FPA算法优化BP神经网络的初始权值和阈值,有效提高了BP神经网络的预报精度和泛化能力,且拓展了BP神经网络初始权值、阈值的优化方法,为相关预报研究提供了参考。     

基于BP-ARIMA的混凝土坝多尺度变形组合预报模型

针对大坝变形常规统计预报模型在监测信息挖掘时的优势单一性及预报精度欠佳等问题,视大坝变形观测资料为非平稳时间序列,从影响大坝变形的因素出发,将其分为周期性影响因素与随机影响因素,利用多尺度小波分析方法将大坝变形监测序列分解并重构,结合BP神经网络与自回归积分滑动平均模型(Autoregressive Integrated Moving Average Model,简记ARIMA)对其随机信号与系统信号分项训练预报,并将其预报值相叠加,据此,应用时间序列原理提出了一种基于BP-ARIMA的混凝土坝多尺度变形组合预报模型。工程实例分析表明,所建组合模型较常规模型能够有效挖掘监测信息中所蕴含的有效成分,预报精度显著提升,且计算分析过程简便,为高边坡及水工建筑物中其他监测指标的预报提供了新方法。     

基于遗传算法优化的小波神经网络在中长期水文预报中的应用——以三门峡为例

造成中长期水文预报研究和预报困难的主要原因是水文现象本身的复杂性和不确定性,以及内部复杂的非线性关系,针对这些问题,本文建立了一种小波神经网络时间序列模型,并用遗传算法对小波神经网络的连接权值和伸缩变量、平移变量进行优化。使用这种模型对黄河三门峡站逐年月天然流量进行预报检验,并将检验结果与传统的小波神经网络和BP神经网络进行对比,从模型预报精度、趋势性和稳定性3个方面进行分析,分析发现,这种遗传算法优化的小波神经网络时间序列模型能够有效地克服传统的小波神经网络和BP网络容易陷入局部极小的缺陷,能够对水文现象的趋势性作出较精确的预测,具有良好的预报精度和稳定性。     

BP神经网络在大坝位移监测中的应用

利用BP神经网络模型较强的非线性处理能力特性,以水位、温度和时效作为输入层,大坝位移为输出层,建立BP神经网络模型对龙滩大坝的位移监测数据进行模拟和预测,并将拟合值、预报值和实测值进行对比分析,结果表明:BP神经网络模型对大坝位移拟合效果较好,预报值精度较高,具有一定的参考应用价值。     

基于LSTM-GM神经网络模型的风暴潮增水预报方法

为充分挖掘风暴潮增水的时序关联特性,提高风暴潮增水的预报精度,综合考虑台风因素、气象要素和天文潮因素对风暴潮增水的影响,结合长短期记忆(LSTM)神经网络和灰色模型(GM)的优势,提出基于LSTM-GM神经网络模型的风暴潮增水预报方法。利用该方法采用12场历史台风数据对小清河入海口风暴潮增水进行模拟预报,并将预报结果与LSTM神经网络、BP神经网络的预报结果进行对比。结果表明：相较于LSTM神经网络和BP神经网络,LSTM-GM神经网络模型的纳什效率系数分别提高了6.5%和11.4%,均方根误差分别降低了70.6%和72.2%,平均相对误差分别降低了50%和69.2%;LSTM-GM神经网络模型可有效处理风暴潮增水与各影响因素间的非线性关系,提高风暴潮增水预报的精度。     

基于MIV的遗传神经网络径流预报模型

针对中长期水文预报中预报对象与预报因子之间复杂的非线性关系,引入平均影响值对预报因子进行筛选,选出对头道拐站年径流量影响较大的年降水量、年均相对湿度、年均气压3个因子作为神经网络的自变量,利用遗传算法优化的BP神经网络建立了预报模型。预报结果表明:基于平均影响值的遗传神经网络的预报精度及稳定性均达到了满意的效果。     

Call for Papers

正As the eldest and most distributed professional journal of water industry,Water Wastewater Engineering(W WE)gained very wide acceptance from the colleagues working in this field in this country since start publication in 1964.Now it is administrated by the Ministry of Construction of the People’s Republic of China and sponsored by the China Civil Engineering Society etc.     

Serial discontinuities in a Rocky mountain river. I. Distribution and abundance of plecoptera

Samples were taken year-round at eleven sites along the altitudinal profile (2900-1400 m a.s.l.) of the Gunnison River, a 329 km tributary of the Colorado River, to document the distribution of the Plecoptera and to evaluate responses to hypolimnial-release dams in the headwaters and middle reaches. Twenty-two species were present, with the greatest species richness occurring in an unregulated segment upstream of the middle reach dams; average nymphal biomass over the study period (175 organisms, 395 mg dry mass m^?2) was also greatest in this segment. Only four species (58 organisms, 48 mg m^?2) were present in the tailwaters of the headwater dam and values were greatly reduced (nine species; 35 organisms, 180 mg m^?2) below the middle reach dams. The stonefly community recovered ca. 80 km downstream from the last dam (15 species; 244 organisms, 250 mg m^?2), apparently in response to natural resetting of environmental conditions corresponding to those above the middle reach dams. At the most downstream site (11) only four species (four organisms, 16 mg m^?2) were present. The observed distributional pattern is a classic serial discontinuity in response to hypolimnial stream regulation in a temperate latitude river.     

地下厂房施工过程的三维非线性数值模拟分析

根据冗各电站地下厂房开挖支护的施工顺序,以隐式杆单元模拟锚杆,以壳单元模拟混凝土喷层,采用三维非线性有限元技术,对洞室群的开挖支护动态过程进行了模拟计算。通过三维非线性有限元计算分析得到:洞室群动态开挖过程中围岩应力变形的分布规律,围岩塑性区的发展规律,锚杆受力变化情况。计算结果表明:整个开挖过程中,洞周围岩位移变化规律正常,量值较小,塑性区主要分布在洞室周边,围岩稳定性较好;但在洞室交叉口处围岩的变形较大,出现较大范围的塑性区,在此基础上提出了支护措施优化方案,并对优化方案的效果进行了评价。     

Impacts of Microcystis on algal biodiversity and use of new technology to remove Microcystis and dissolved nutrients

Cyanobacteria often dominate eutrophic lakes, outcompeting green algae that are required by fish and zooplankton. This study was undertaken to ascertain the impact of the cyanobacterium, Microcystis, on algal biodiversity. Under laboratory conditions, we found that the presence of Microcystis decreased phyla richness by 58%, phyla evenness by 47%, genera richness by 66% and genera evenness by 51%. Analysis by mixed anova s demonstrated a significant interaction between treatment and time and confirmed a significant reduction in richness and evenness of phyla and genera. We also conducted a phosphate restriction assay on the algae in Mason Lake (Irvine, CA, USA) and found that the threshold needed for algal growth there was 0.02 mg L^?1 PO₄. A pilot study was then conducted to test the effectiveness of the Blue Pro? water treatment facility in removal of this colonial organism from Mason Lake, in addition to removal of dissolved nutrients required for its growth. We measured a 97% reduction in Microcystis cells, a 72% reduction in chlorophyll‐a, and a 96% reduction in phosphate after just one 10 min cycle through the unit. Our study demonstrated that removal of Microcystis colonies may allow green algae to increase in numbers. This may improve algal biodiversity, which will benefit zooplankton and fishes.     

Evaluation of freshwater mussel relocation as a conservation and management strategy

The relocation of unionacean mussels is commonly used as a conservation and management tool in large rivers and streams. Relocation has been used to recolonize areas where mussel populations have been eliminated by prior pollution events, to remove mussels from construction zones and to re-establish populations of endangered species. More recently, relocation has been used to protect native freshwater mussels from colonization by the exotic zebra mussel Dreissena polymorpha. We conducted a literature review of mussel relocations and evaluated their relative success as a conservation and management strategy. We found that 43% of all relocations were conducted because of construction projects that were forced to comply with the Endangered Species Act 1973 and that only 16% were monitored for five or more consecutive years. Most (43%) relocation projects were conducted from July to September, presumably a period when reproductive stress is relatively low for most species and the metabolic rate is sufficient for reburrowing in the substrate. The mortality of relocated mussels was unreported in 27% of projects; reported mortality varied widely among projects and species and was difficult to assess. The mean mortality of relocated mussels was 49% based on an average recovery rate of 43%. There is little guidance on the methods for relocation or for monitoring the subsequent long-term status of relocated mussels. Based on this evaluation, research is needed to develop criteria for selecting a suitable relocation site and to establish appropriate methods and guidelines for conducting relocation projects.     

基于模型试验的河床砂砾石层基本特性研究

为了经济可靠地确定深厚覆盖层的基本特性参数,以大渡河双江口水电站坝基河床砂砾石层基本特性研究为工程背景,进行大尺寸模型试验,研究河床砂砾石层的旁压模量和动探击数随密度、级配以及上覆压力的变化规律。根据室内模型试验结果,结合现场旁压试验及钻孔试验成果,推断出坝址区河床砂砾石相关层位的密度,并依此进行砂砾石室内力学试验,获得设计所需力学参数。研究成果可为深厚覆盖层基本特性测试提供新的方法。     

SPATIAL AND TEMPORAL WATER QUALITY VARIABILITY IN AQUATIC HABITATS OF A CULTIVATED FLOODPLAIN

The floodplains of lowland rivers contain diverse aquatic habitats that provide valuable ecosystem services but are perturbed when intensively cultivated. Hydrologic, water chemistry and biological (fish) conditions in five aquatic habitats along the Coldwater River, Mississippi, were measured for more than 4 years: the river, two severed meanders that functioned as backwaters, a managed wetland and an ephemeral channel draining cultivated fields. Off‐channel habitats were connected to downstream regions 0.10% to 32% of the dry season and 24% to 67% of the wet season. The median temperatures for the five monitored sites ranged from 18°C to 23°C, the median total solids concentration for all sites was 135 mg L^?1, the median total phosphorus was 0.29 mg L^?1 and the median total nitrogen was 1.56 mg L^?1. Chemical and physical water quality displayed strong seasonal differences between the wet winter/spring and the dry summer/fall periods so that temporal variation consisted of gradual seasonal trends superimposed on strong diurnal variations. All off‐channel habitats exhibited periods of hypoxia and temperatures >30°C during the dry season. Between‐site gradients of water and habitat quality were strongly coupled to water depth and runoff loading. The rehabilitation of one backwater by increasing water depth and diverting agricultural runoff was associated with improved water quality and fish species richness relative to an adjacent untreated backwater. The diversion of polluted runoff and the use of water control structures to maintain greater water depth were observed to be effective management tools, but the former reduces the water supply to habitats that tend to dry up and the latter reduces connectivity. Published in 2011 by John Wiley & Sons, Ltd.     

簸箕李引黄灌区水沙分布及优化调度的经验

灌区利用水沙观测所取得的资料和试验数据，运用水沙运动力学理论，总结分析和揭示了灌区水沙分布的规律和特点。在此基础上，结合灌区的实际运行情况，围绕减少骨干渠道的淤积进行探讨、研究，进而，优化水沙调度方案。该方案实施后，骨干渠道减旅伴效果明显，取得了成功的经验。     

小湾水库大坝防渗面板及其裂缝的加固处理

小湾水库大坝防渗面板整体抗渗性能差,且存在多条裂缝,通过对面板钢丝网喷涂丙乳砂浆加固处理及其裂缝修补,大坝渗漏量明显减少,且左岸上灌浆廊道内及大坝下游坝面92～93 m高程下廊道入口处原有小股射流全部消失.     

Spatio‐temporal variation of gross CO2 and CH4 diffusive emissions from Australian reservoirs and natural aquatic ecosystems,and estimation of net reservoir emissions

Carbon dioxide (CO₂) and methane (CH₄) diffusive emissions were measured during two field surveys in Queensland and Tasmania, Australia, using the floating chamber method. Bubbling and degassing emissions in 2010 were estimated in Koombooloomba Dam reservoir using only inverted funnels and gas concentrations, respectively. A total of 14 reservoirs and 16 rivers and lakes were sampled from 2006 to 2010. Spatial variation was substantial within each water body, as well as between them. The main drivers of diffusive emission variation were physiographic region and climate, with a clear demarcation being observed between diffusive emissions from tropical Queensland and temperate Tasmania, and between the humid West Coast Range (Tasmania) and dry Central Plateau (Tasmania). Higher CO₂ and CH₄ diffusive emissions were observed during the dry season, when long water residence times would promote organic matter degradation. Estimated total gross emissions, including diffusive, bubbling and degassing emissions, for Koombooloomba Dam reservoir were about 1.5 × 10⁶ t CO₂eq km² per year, or 24 × 10⁶ t CO₂eq per year. This corresponds to a plant emission factor of 3.18 kg CO₂eq MWh^?1. Using an estimate of terrestrial emissions derived from literature data for the Tully River catchment area, rough estimated net emissions from the catchment area are about 44 kt CO₂eq per year, or 5.83 kg CO₂eq MWh^?1, which is in the lower range of the studied reservoirs.