基于支持向量机的水电站中长期径流组合预报

中长期径流预报是充分利用水资源、实现水电站优化运行的重要环节。以金沙江中游龙盘电站为研究对象,分别采用自回归滑动平均模型、最近邻抽样回归模型、BP神经网络建立了该水电站月径流预报模型,在分析三种模型预报结果具有一定互补性的基础上进一步建立了支持向量机分月组合预报模型。统计结果表明,与单一预报模型相比,该组合预报模型具有更高的精度和稳定性,为寻求水电站径流预报规律和制定中长期调度计划提供了技术支持。     

福建省电网中长期水文预报系统设计与开发

利用Java技术平台,采用双机群集数据库系统(Orade 10g)成功设计并开发了基于Web的福建省电网中长期水文预报系统.将长期预报和中期预报分离,前者采用多种数理统计模型,以月或旬为预报时段提供未来1 a的径流预报及相关的频率预报;后者采用考虑城市气象预报的人工神经网络模型,以日为预报时段提供未来7 d的自动、滚动式径流预报.重点阐述了系统架构设计、业务逻辑设计、功能模块设计和程序开发过程中的关键技术问题.     

GFS可利用性研究及其在旬径流预报中的应用

分析了中长期降雨径流预报研究应用现状和存在的主要问题,研究了利用GFS降雨数值预报信息进行旬径流预报的方法,建立了旬降雨-旬初土壤含水量-旬径流预报模型(PSRM).通过对实例流域历史旬径流资料的模拟优选出模型参数,分析了GFS降雨数值预报信息的可利用性,并给出了利用GFS降雨预报信息和PSRM模型推求径流的方法.结果表明,基于GFS预报降雨信息和PSRM模型推求径流期望值和概率分布用于水库的实时调度可行.     

基于遗传程序设计的中长期径流预报模型研究与应用

应用遗传程序设计建立径流中长期预报模型,结合径流序列数据的特点通过自相关分析确定其滞时输入变量的个数,采用均方误差作为其适应度评价函数,以漫湾实测月径流序列(1953~2003年)和洪家渡实测月径流序列(1951~2004年)为例,通过与ARMA模型、人工神经网络模型的预报结果比较,显示该模型应用于径流中长期预报简单易行且精度较高。     

BP神经网络模型结构对漫湾径流预报精度的影响研究

以云南省漫湾水电站历史径流状况为研究对象，运用三层前馈反向传播神经网络模型对径流进行中长期预报。为解决神经网络预报模型结构难以确定的问题，尝试在预报过程中通过改变该网络模型的结构并对得到的结果进行比较，从而找到适合该径流序列的最佳神经网络模型结构。实际应用表明，使用该结构的模型在实际预报过程中取得了良好的效果。     

GFS预报信息在水电站运行中的应用研究

针对水电站水库的中长期优化调度中若能充分有效地利用预报信息将可获得更大经济效益的问题,利用GFS发布的10~14 d降雨数值预报信息,构建了其信息的统计方法,并以预报降雨为主因子建立了二滩水电站汛期旬径流的枯、平、丰三级预报模型,基于已建立的二滩水电站有、无预报信息相结合的发电调度模型,针对汛期无径流预报的时段,利用径流定性预报结果修正改进了其调度决策。模拟调度结果表明,改进方案可行、优越。     

三峡水库中长期径流预报方法研究

中长期径流预报是充分利用水资源、发挥电站经济效益的有力手段。以三峡水库为研究对象,分别采用周期外延叠加技术、人工神经网络模型、投影寻踪自回归模型和支持向量机回归模型对三峡水库逐月入库径流进行预报。从不同侧面比较分析了这四种方法优劣,并总结各预报模型计算结果的特征及规律,为三峡水库寻求径流预报规律和制定未来中长期调度计划提供了技术支持。     

水电站水库预报发电调度的模糊风险分析

针对径流预报误差给水电站水库发电调度带来的风险,在采用组合模型对月径流进行预报和优化调度计算的基础上,构建了水电站水库预报发电调度的模糊风险分析模型框架,并在对入库径流进行随机模拟的条件下设计了模糊风险分析模型的求解流程。三峡水电站水库的实例应用表明,预报误差在5、10月对发电调度的影响较大,且水电站在枯水期的发电模糊风险比汛期高。     

基于Box-Cox算法的中期径流贝叶斯概率预报方法

为进一步降低中期径流预报信息受降雨不确定性和模型误差等因素的影响,采用贝叶斯预报处理器(BPF)对径流自身和预报过程的不确定性进行耦合,建立中期径流概率预报模型,对中期径流预报的不确定性进行描述。首先将Box-Cox正态变换引入BPF模型,建立基于Box-Cox变换的BPF模型;然后采用逐次优化算法(POA)对各时段的变换系数进行优化,并基于优化后的变换系数建立新的BPF模型;最后对变换系数优化前和优化后建立的BPF模型预报结果进行对比分析。实例应用表明,基于Box-Cox变换的BPF模型可较好地描述预报信息不确定性;变换系数经POA优化后,可进一步提高BPF模型的预报精度。研究成果可为中长期径流预报提供参考。     

模式识别方法在径流长期预报中的应用

预测水电站所在河流未来的径流情势,可为水电站合理运行提供科学的依据,对实现预期的各项经济效益指标有着相当重要的意义.本文应用模式识别的基本原理和方法,对龙羊峡电站入库径流的长期预报进行了尝试,得到了较为满意的结果.     

Should we build more large dams? The actual costs of hydropower megaproject development

A brisk building boom of hydropower mega-dams is underway from China to Brazil. Whether benefits of new dams will outweigh costs remains unresolved despite contentious debates. We investigate this question with the “outside view” or “reference class forecasting” based on literature on decision-making under uncertainty in psychology. We find overwhelming evidence that budgets are systematically biased below actual costs of large hydropower dams—excluding inflation, substantial debt servicing, environmental, and social costs. Using the largest and most reliable reference data of its kind and multilevel statistical techniques applied to large dams for the first time, we were successful in fitting parsimonious models to predict cost and schedule overruns. The outside view suggests that in most countries large hydropower dams will be too costly in absolute terms and take too long to build to deliver a positive risk-adjusted return unless suitable risk management measures outlined in this paper can be affordably provided. Policymakers, particularly in developing countries, are advised to prefer agile energy alternatives that can be built over shorter time horizons to energy megaprojects.     

二滩水电站中长期径流预报研究

针对二滩水电站的实际径流特性和水电站发电调度的要求,应用季节性自回归模型和人工神经网络模型对二滩水电站的月径流、汛期分段和年径流预报进行研究.结果表明,这两种模型对二滩水电站的月径流预报、汛期定性预报均达到了一定精度,可为二滩水电站优化调度的径流输入提供参考依据,尤其是AR(P)模型的非汛期月径流预测和BP模型年径流预测结果可在实际运行中使用.     

Hydropower externalities: A meta-analysis

This paper presents a meta-analysis of existing research related to the economic valuation of the external effects of hydropower. A database consisting of 81 observations derived from 29 studies valuing the non-market impacts of hydropower electricity generation is constructed with the main aim to quantify and explain the economic values for positive and negative hydropower externalities. Different meta-regression model specifications are used to test the robustness of significant determinants of non-market values, including different types of hydropower impacts. The explanatory and predictive power of the estimated models is relatively high. Whilst controlling for sample and study characteristics, we find significant evidence for public aversion towards deteriorations of landscape, vegetation and wildlife caused by hydropower projects. There is however only weak evidence of willingness to pay for mitigating these effects. The main positive externality of hydropower generation, the avoidance of greenhouse gas emission, positively influences welfare estimates when combined with the share of hydropower in national energy production. Sensitivity to scope is detected, but not linked to specific externalities or non-market valuation methods.     

三维可视化技术在水电站设备管理中的应用研究

水轮机、发电机、主变压器是抽水蓄能电站的重要生产设备,担负着水电站发电和输电生产过程,无论从设备资产的占有率,还是从管理工作的内容本身,生产设备的稳定运行关系到生产秩序的正常开展.研究采用基于二三维相结合的可视化技术[1],构建水电站设备三维模型,并将设备模型精细构建到零部件级别;以模型为载体,一方面挂接设备从安装、运行、维护的全过程资料,另一方面与HPMS管理系统中设备树的1~5级进行数据深度集成,将设备实时运行数据叠加到模型及零部件上,通过三维可视化技术超现实展示方式创新性建立了水电站三维全息化设备管理平台[2],以三维可视化方式实时监控设备及零部件运行状态,提高设备管理水平.     

岷江流域某小型水电站灾后监控系统改造

四川岷江流域某小型水电站连续遭受两次大型自然灾害,灾后监控系统出现机组供水、油压监控准确性降低,球阀、高压气机、主变保护监控功能缺失等问题。对此,结合电站实际情况及受灾类型,以PLC为基础、以提高监控效率为目标,完善了机组供水、主变保护、机组油压监控装置,对球阀监视装置、高压气机控制装置等进行了升级改造。实践证明,监控系统稳定性、监控范围及效率得以提高,误报率降低,为中小型水电厂监控系统的升级改造提供了参考。     

基于中尺度气候模式的降水对水力发电的影响

通过分析水力发电站发电量与地区降水量之间的关系建立由降水驱动的水电响应模型,并在此基础上,使用逐步聚类方法对中尺度的区域气候模式进行统计降尺度,得到未来高精度的降水数据,以此研究未来降水变化对水力发电的影响。以大渡河流域某电站为案例,对其发电量进行模拟与预报,研究在未来气候变化条件下该电站的水力发电变化。结果表明,该电站在未来不同月份和调度时期的降水变化存在差异,由此会造成未来发电高峰期缩短,发电峰值下降,但全年总发电量基本保持不变。     

An estimation of the Swiss hydropower rent

The electricity generation in Switzerland is mostly based on hydropower (∼58%) and nuclear power (∼38%). The exploitation of water in the hydropower sector can generate significant economic rent. One possibility to capture this rent is through royalties or fees. This system has been used in Switzerland since many decades. However, the actual water fee system is not flexible and does not take into consideration different production costs between the type and location of hydropower plants. Furthermore, storage plants can sell electricity to a higher price than run-of-river plants. A flexible system is needed in a liberalized electricity market, to take into account these different production situations and the fact that prices may vary considerably over time. The main goal of this paper is to calculate the potential economic rent that could be generated in the Swiss hydropower sector under a future liberalized market. Based on the results of the paper, it can be concluded that the introduction of a flexible fee system could improve the competitiveness of the hydropower sector and promote an energy system based on renewable energy sources.     

梯级水电站群超短期入库流量滞时时间序列预报方法

超短期预报调度对于实现短期调度和实时调度无缝衔接、降低调度风险、提高水电调度水平意义重大。而实际超短期预报调度面临着有效预报信息短缺、实时采集信息错报漏报、支流小水电调节、模型通用性差等一系列问题,为此,提出了一种梯级水电站群超短期滞时时间序列预报方法,首先分析了上游电站历史出库和下游电站历史入库流量,利用相关系数找出最强滞时流量匹配关系,从而将上游电站出库流量从下游电站入库中分解出来,还原出有效的区间流量,再通过区间流量逐日化进行时间序列建模,最后与上游电站滞时出库合成为下游电站预报入库流量。以云南澜沧江干流梯级电站为例,取得了良好的预测效果,从而验证了该方法的合理性、准确性与通用性。     

黑河流域梯级水电站群机组远程监控与故障诊断系统

为了确保水轮发电机组的安全稳定和经济运行,提出了基于Internet/Intranet开放式结构的水轮发电机组远程监测与故障诊断系统,并对系统的设计原则、网络化体系结构、机组测点布置、传感器选型及系统的功能做了详细的分析。该系统实现了水轮发电机组信号的数据采集、数据分析和处理,并能通过Internet/Intranet进行故障诊断。实际应用表明,该系统兼具技术的先进性和实用性,运行稳定可靠,具有良好的应用前景。     

Life cycle assessment of a community hydroelectric power system in rural Thailand

Rural electrification and the provision of low cost, low emission technology in developing countries require decision makers to be well informed on the costs, appropriateness and environmental credentials of all available options. While cost and appropriateness are often shaped by observable local considerations, environmental considerations are increasingly influenced by global concerns which are more difficult to identify and convey to all stakeholders.Life cycle assessment is an iterative process used to analyse a product or system. This study iteratively applies life cycle assessment (LCA) to a 3 kW community hydroelectric system located in Huai Kra Thing (HKT) village in rural Thailand. The cradle to grave analysis models the hydropower scheme’s construction, operation and end of life phases over a period of twenty years and includes all relevant equipment, materials and transportation.The study results in the enumeration of the environmental credentials of the HKT hydropower system and highlights the need to place environmental performance, and LCA itself, in a proper context. In the broadest sense, LCA results for the HKT hydropower system are found to reflect a common trend reported in hydropower LCA literature, namely that smaller hydropower systems have a greater environmentally impact per kWh – perform less well environmentally - than larger systems. Placed within a rural electrification context, however, the HKT hydropower system yields better environmental and financial outcomes than diesel generator and grid connection alternatives.