【作者】 段富；

【导师】 谢克明；

【作者基本信息】 太原理工大学 ， 电路与系统， 2010， 博士

【摘要】 随着全球人口与经济的快速增长以及气候的变化,各地区水资源分布不均匀以及水资源短缺的现象,已成为经济和社会发展中急需解决的问题之一。兴建跨流域或跨地区的大型梯级引水工程,成为缓和以至解决这一问题的有效途径。大型梯级引水工程是一项复杂的系统工程,对这类工程自主优化调度的研究成为智能优化调度领域面临的重要课题。论文以万家寨引黄工程为背景,基于自主计算技术的理论和智能优化调度方法,在大型梯级引水工程SCADA系统和仿真系统的基础上,对大型梯级引水工程的自主优化调度模型及其仿真进行研究,建立了满足供水需求的水库长期优化调度数学模型以及梯级引水泵站的短期优化调度数学模型,设计了应用人工免疫系统求解模型的算法,并进行仿真实验。最后,提出大型梯级引水工程自主优化调度系统的框架及其形式化描述。论文研究内容融合信息科学、计算机科学、自动化科学、水力学仿真、以及管理科学等多个学科的交叉领域。主要创新性工作包括：(1)依据万家寨引黄工程工程监控和调度的特点及要求,提出了集中统一调度、功能分层分散控制的SCADA系统模式,并运用于引黄工程SCADA系统设计中,可供大型梯级引水工程SCADA系统鉴戒。(2)提出了依据引水工程运行调度、流域的径流预测和供水区域的需水情况,以全线输水耗能最少为目标的全线自主优化调度模型的基本框架及其形式化描述,能够根据流域的径流和供水区域的需水变化,在领域知识指导下,自主地进行优化调度。(3)针对供水水库的径流以及供水需求,建立了水库长期优化调度的数学模型、BP神经网络的径流预测模型和水库优化调度知识库,提出了知识导向的求解水库优化调度模型的改进免疫规划算法。(4)推导出了梯级输水泵站优化调度问题的分段水力学模型,按照“大系统分解-协调法”,设计了分层分段的优化调度模型及其分层克隆选择优化算法(HCSA)。更多还原

【Abstract】 With climate change and highly-speed development of world economy, the uneven and shortage appearance of water recourse in districts has become urgent problem in society development. Building large-scale multi-stage water transfer project, which span valleys and districts, has become an effective way to tackle it. For it is a sophisticated project, the research of its autonomic optimization scheduling has become an important subject in the field of intelligence optimization.Based on autonomic computing theory and intelligence optimization scheduling method, the paper takes Wanjiazhai Yellow River Diversion Project as background and has research on autonomic optimization scheduling model of large-scale multi-stage water transfer project, which is on the basis of SCSDA system and Hydraulic simulation system. It not only constructs long-term optimization scheduling mathematic model of storage reservoir, but also short-term mathematic model of multi-stage priming pump station. Besides it, the paper designs the algorithm of solving the model with artificial immune and take simulation. Finally, it puts forward the frame and the formalization description of autonomic optimization scheduling system of large-scale multi-stage water transfer project. The research involves such subject as information science, computer science, automatic science, hydraulics simulation and management science. The innovative achievements of the paper can be concluded as following.(1)According to the character and demand of supervision and scheduling of the project, it proposes SCADA system pattern based on centralized dispatcher and decentralized control, It was applied into the design of SCADA system of the project, which provides new clues to SCADA system of large-scale multi-stage water transfer project.(2)According to run schedule of water transfer project, runoff forecast and demand of water supply, it proposes the basic frame and formalization description of autonomic optimization scheduling model, which minimize the cost in water delivery along the line. Based on runoff and change of water demand, it can optimize the scheduling independently.(3)Aimed at runoff of storage reservoir and supply demand, it constructs mathematic model, runoff forecast model based on BP neural network and scheduling knowledge base. It proposes improved immune programming algorithm to solve optimization scheduling model.(4)It deduces subsection hydraulics model of multi-stage water delivery pump station optimization. According to large-scale decomposition-coordination method, it designs optimization scheduling model and its hierarchy clone selection algorithm (HCSA).更多还原