丁东水库提水调度及健康评价研究

发布时间：2019-04-03 07:29

【摘要】：水资源是经济社会发展的重要物质基础,是一个国家最重要的战略资源之一。我国水资源人均不足且时空分布不均,这给我国经济发展和社会进步带来很大的挑战。在平原地区,解决此问题的重要措施之一为在水资源缺乏地区修建一定数量的平原水库,通过其对水的蓄、排,实现水资源在时空上再分配,能够有效缓解日趋严峻的水资源供需矛盾。山东省平原水库建设起步早、发展快,但是在其运行管理中存在某些不足。本文通过对丁东水库的分析发现其不足之处主要为:一是平原水库的运行调度以经验调度为主,调节功能较为低下,造成无法达到供水对象设计保证率的要求;二是修建过程中受当时科学技术条件的限制,存在着设计标准偏低、施工工序存在不足以及缺乏对生态环境考虑等问题,致使平原水库运行管理中存在风险,缺乏完善的评定平原水库健康状态的评价体系。本文针对上述两个问题开展了研究,主要研究内容如下:(1)分析影响平原水库提水量的主要因素,以入库泵站运行周期能源单耗最小为目标函数,使用遗传算法对历史资料进行优化调度得到水库提水调度数据集,采用能够将平原水库当前蓄水情况、气象预报信息、水库多年的实际运行情况与调度者经验相结合的决策树C4.5算法从中提取提水调度规则决策树。利用决策树C4.5算法在丁东水库2002~2010年提水调度数据集中提取调度规则决策树,采用2011~2015年数据进行验证,结果证明相较于实际调度,前者具有提高供水保证率和泵站效率以及减少能源消耗的优点,相较于遗传算法具有操作性强的优势。(2)综合考虑影响平原水库健康的各方面因素,提出了健康平原水库的定义;然后在此基础上建立了平原水库健康评价体系,该体系包括结构安全、功能效益、生态环境和健康稳定性4个一级指标以及14个二级指标,按照健康、亚健康、不健康、病态4个级别进行划分,并确定了各指标标准特征值。针对平原水库健康评价指标的模糊特性,选用可变模糊方法建立了平原水库健康评价模型,并采用综合赋权法计算各指标的权重。通过对2011~2015年丁东水库健康状况的评价表明本文构建的平原水库健康评价体系考虑的因素全面,评定等级划分合理,划分依据规范,可变模糊评价方法能够很好地处理评价指标的模糊特性,并且评价结果准确、操作性强。
[Abstract]:Water resources are the important material basis of economic and social development and one of the most important strategic resources of a country. The shortage of water resources per capita and the uneven distribution of time and space in China have brought great challenges to the economic development and social progress of our country. In plain areas, one of the important measures to solve this problem is to build a certain number of plain reservoirs in areas where water resources are scarce, so as to realize the reallocation of water resources in time and space through the storage and discharge of water resources. It can effectively alleviate the increasingly severe contradiction between supply and demand of water resources. The plain reservoir construction in Shandong Province started early and developed rapidly, but there are some deficiencies in its operation and management. Based on the analysis of Dingdong reservoir, it is found that the deficiency is as follows: first, the operation and operation of plain reservoir is mainly based on empirical regulation, and the regulation function is relatively low, which makes it impossible to meet the requirement of guarantee rate of water supply design; Second, because of the limitation of science and technology conditions at that time, there are some problems such as low design standard, insufficient construction procedure and lack of consideration of ecological environment, which lead to risks in the operation and management of plain reservoir. There is no perfect evaluation system for evaluating the health status of plain reservoirs. The main contents of this paper are as follows: (1) the main factors affecting the water yield of plain reservoir are analyzed, and the objective function is the minimum unit energy consumption in the operation period of the pumping station in the reservoir. The genetic algorithm is used to optimize the operation of historical data to obtain the data set of reservoir water lifting and dispatching. The current water storage situation and meteorological forecast information of plain reservoir are adopted. The decision tree C4.5 algorithm, which combines the actual operation of the reservoir for many years and the dispatcher's experience, extracts the decision tree of the water extraction regulation from the decision tree. The decision tree C4.5 algorithm is used to extract the decision tree of regulation rules in Dingdong Reservoir from 2002 to 2010. The results show that compared with the actual operation, the decision tree is verified by the data from 2011 to 2015. The former has the advantages of improving water supply guarantee rate, pumping station efficiency and reducing energy consumption, and has strong operational advantages over genetic algorithm. (2) the factors that affect the health of plain reservoir are considered comprehensively, and the former has the advantages of improving water supply guarantee rate and pumping station efficiency and reducing energy consumption. The definition of healthy plain reservoir is put forward. Then, the health evaluation system of plain reservoir is established. The system includes four first-order indexes: structural safety, functional benefit, ecological environment and health stability, and 14 secondary indexes, according to health, sub-health and unhealthy. The ill-conditioned four grades were divided and the standard eigenvalues of each index were determined. According to the fuzzy characteristic of plain reservoir health evaluation index, the variable fuzzy method is used to establish the plain reservoir health evaluation model, and the weight of each index is calculated by comprehensive weighting method. Through the evaluation of the health status of Dingdong Reservoir from 2011 to 2015, it is shown that the health evaluation system of plain reservoir constructed in this paper takes into account all the factors, the classification of assessment grades is reasonable, and the criteria for classification are reasonable. The variable fuzzy evaluation method can well deal with the fuzzy characteristics of the evaluation index, and the evaluation results are accurate and operational.
【学位授予单位】：山东农业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TV697.1

【参考文献】