节点文献

双标准的水库防洪预报调度方式研究及其应用

Research and Application on the Double Standard Flood Control Forecast Operation Mode

【作者】 刁艳芳

【导师】 王本德;

【作者基本信息】 大连理工大学 , 水文学及水资源, 2010, 博士

【摘要】 水库作为重要的防洪工程措施,在洪水资源化方面发挥了重要的作用。实践证明,科学的水库防洪调度设计是安全经济建设水库与实时调度发挥设计效益的关键。不考虑预报设计的调度方式往往贻误泄流或蓄水时机,增加防洪风险或造成洪水资源的浪费。为了增加洪水资源可利用量、充分发挥水库防洪兴利综合效益,近20年我国在已建水库除研究了汛限水位动态控制理论与方法外,重视研究防洪预报调度方式,如考虑洪水预报或二级降雨预报信息的预报调度的方式,分类防洪预报调度方式等。但是这些方法没有考虑水库设计洪水特性及水库的调节能力,仅仅考虑洪水预报精度及其风险,因此深入研究与完善预报调度方式设计理论与方法,既是充分挖掘水库防洪、兴利综合效益潜力的要求,也是水库防洪调度理论研究自身发展的必然趋势。本文将最大熵原理、信息熵、随机模拟技术、模糊优选模型等理论融合入水库防洪预报调度方式的设计理论与方法中,系统研究了洪水预报误差特性、双标准的水库防洪预报调度方式设计方法及风险分析和双标准预报调度规则的水库实时调度方案优选方法。主要研究内容和成果如下:(1)针对目前洪水预报误差分布规律研究较少和观点不一的问题,本文建立了洪水预报误差分布的最大熵模型,分析了我国湿润和半湿润地区九座水库的洪水预报误差,初步总结出我国湿润和半湿润地区的洪水预报误差近似服从正态分布。(2)为了将预报方案安全、经济的用于防洪预报调度之中,本文提出了洪水预报误差域、水库防洪预报调度可承受误差域及其可靠度的概念和数值求解方法,并以桓仁水库和白龟山水库为例进行了计算,总结出可承受误差域及其可靠度的基本特性,并为制定双标准预报调度规则奠定了基础。(3)目前水库实施预报调度规则设计的基本条件是洪水预报方案精度要达到规范的标准,尚未考虑水库设计洪水、调节特性及其上下游防洪能力的安全约束,缺乏水库安全调度的可靠度标准的问题,为此本文提出“预报方案的规范评定等级和水库可承受误差域的可靠度”双标准设计水库防洪预报调度规则的新方法。结合桓仁水库和白龟山水库的实例探讨了此方法的可行性。(4)针对已有的防洪预报调度方式风险分析考虑风险因素较少的问题,提出了水库防洪预报调度方式的综合风险分析模型。重点分析了水文、水力、水位-库容关系和调度滞时4种不确定性因素产生的原因及其概率分布,建立了基于蒙特卡洛随机模拟的水库防洪预报调度方式综合风险分析模型,此模型适用于单因素和多因素共同影响下的风险分析体系。并以桓仁水库和白龟山水库为例,计算了单一标准和双标准预报调度方式的风险率。结果表明,水文不确定性在大多数情况下是最主要的风险源,并且双标准预报调度方式比常规调度方式、单一标准预报调度方式对水库及其下游更为安全。(5)为了将双标准预报调度规则研究成果与防洪调度系统实时调度相匹配,本文提出了基于双标准预报调度规则的水库实时调度方案优选方法。在此方法中,双标准预报调度规则作为第一可行调度方案,交互生成其它方案,选取水库调洪最高水位、下游最大组合流量、发电量、弃水量和调洪末水位作为优选指标,采用改进的熵权法、二元对比法和基于博弈论的组合赋权法确定指标权重,并使用多层多目标模糊优选模型优选满意调度方案。同时设计了基于双标准预报调度规则的人机交互实时调度方案优选决策支持系统,可方便快捷的优选实时调度方案。最后对全文作以总结,并对有待进一步研究的问题进行了展望。

【Abstract】 Reservoirs as an important flood control structural measure play an important role in flood resources utilization. It has been found that the scientific design of reservoir flood control operation is the key to construct reservoirs safely and economically and play designing benefit in real-time operation. Reservoir flood control operation mode without forecast information always delays the opportunity to release or impound, increases flood risk and wastes flood resources. To increase flood resource volumes and improve flood control and utilizable benefits, the flood control forecast operation mode is studied in the past two decades except the theory and method of dynamic control of the limited water level, which involves forecast operation mode based on flood forecasting or two-stage rainfall forecasting and classified forecast operation mode etc. While the absence of considering characteristics of design flood hydrographs and regulate performance of reservoirs, these methods only consider the flood forecast accuracy and risk. So the research and supplement of the theory and method of the forecast operation mode is not only the demand for flood control and beniefit, but also the development trend of the reservoir flood control operation theory. In this dissertation, the methods including the principle of maximum entropy, information entropy, stochastic simulation technique and fuzzy optimization model etc. are employed into the theory and method of the reservoir flood control forecast operation mode. The characteristics of flood forecast errors, designing method and risk analysis of double standard flood control forecast operation mode and a real-time operation optimization model based on double standard flood control forecast operation rule are studied. The main researches and results are as follows:(1) At present, researches on the distribution of flood forecast error are very few and inconsistent views. The flood forecast error distribution model based on the principle of maximum entropy is established, and the flood forecast errors of nine reservoirs in humid and sub-humid regions are analyzed by proposed model. The results show that the flood forecast error obeys the normal distribution approximately in humid and semi-humid regions.(2) To employ flood forecast in flood control forecast operation safely and economically, this dissertation puts forward the definitions and numerical solutions of the flood forecast error domain, the reservoir flood control forecast operation acceptable error domain and its reliability. Taking Huanren and Baiguishan reservoirs as examples, the general rules of the reservoir flood control forecast operation acceptable error domain and its reliability are summarized which are the foundations for the design of the double standard flood control forecast operation rule.(3) The basic condition of designing the flood control forecast operation rule is the forecast accuracy achieving the published standard according to the "Standard for hydrological information and hydrological forecasting SL250-2000". That is single standard method, which cannot consider the design flood hydrographs and the regulating characteristics of the reservoir and the security constraints of the upstream and downstream flood control capacities. Moreover, this method lacks the reliability standard of the safe flood control operation. Therefore, the double standard flood control forecast operation rule is employed by combining the forecast accuracy and the reliability of the acceptable error domain. This method is proved to be feasible by two examples of Huanren and Baiguishan reservoirs.(4) The problem of existing risk analysis on flood control forecast operation mode is that they consider a few of risk factors. So, this dissertation proposes integrated risk analysis model of flood control forecast operation mode. Causes and probability distributions of four uncertainties including hydrological uncertainty, hydraulic uncertainty, stage-storage uncertainty and time-delay uncertainty are analyzed. And integrated risk analysis model of flood control forecast operation mode is presented based on Monte Carlo stochastic simulation theory, and this model is appropriate to single and combined factors. Taking Huanren and Baiguishan reservoirs as examples, risk rates of single and double standard flood control forecast operation modes are calculated. The results show that hydrological uncertainty is mainly risk source in most cases, and the double standard forecast operation mode is safe to reservoirs and their lower reaches than conventional operation mode and the single standard flood forecast operation mode.(5) In order to combine the double standard forecast operation rule and real-time flood control operation system, a real-time operation optimization model based on double standard flood control forecast operation rule is established. In this model, the double standard forecast operation rule regards as the first feasible operation scheme, and other schemes are establish by human-computer interaction. The highest water level, peak discharge, generating capacity, waste water volume and terminal water level are selected as indices, and modified entropy method, method of two elements contrasting and combination weighting approach based on game theory are adopted to calculate weight. In the end, fuzzy optimization model is introduced to optimize schemes. At the same time, decision support system of a real-time operation optimization model based on double standard flood control forecast operation rule is proposed. This decision support system could optimize schemes quickly. Lastly, the conclusions and problems need to be further studied are given at the end of the dissertation.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络