梯级水电站群联合优化调度及其决策方法

发布时间：2018-03-14 00:39

  本文选题：梯级水电站　切入点：多目标优化调度　出处：《华北电力大学》2014年硕士论文　论文类型：学位论文

【摘要】：大型梯级水电站具有复杂的流域拓扑结构、紧密的电力和水力联系等特点。由于上述特点导致了其联合优化调度较为复杂；同时,大型梯级水电站具有大库容、高水头的特征,优化时还存在大范围寻优而导致计算效率低下的问题。除此之外,电网负荷与梯级水电站入库径流的不确定性,对水电站安全、稳定、经济运行带来较大的负面影响。因此,亟需开展大型梯级水电站联合优化调度研究,尤其是短期优化调度研究,以快速响应电网负荷与梯级水电站入库径流的突变,提高水头效益,增加水能利用率。在确保梯级水电站群安全性的前提下,实现经济效益、通航率最大化。 本文在深入分析了传统优化方法在解决梯级水电站联合优化调度问题时,存在“维数灾”、寻优效果不理想等不足的基础上,利用系统工程理论和现代智能优化算法,对多目标优化算法、多属性决策方法及其在梯级调度决策中的应用进行了研究。论文具体内容概括如下： 本文首先根据三峡-葛洲坝实际运行数据和工况,基于最小二乘法,建立了三峡-葛洲坝梯级水电站的水力计算流程。通过对基本数和水力计算流程的研究分析,建立了三峡-葛洲坝梯级水电站短期优化调度数学模型。为智能优化算法的使用创造了条件；其次基于建立的短期优化调度数学模型,运用权重法将多目标的优化调度问题转化为单目标问题,并且采用基于Sigmoid曲线改进后的自适应遗传算法进行模型求解。根据三峡-葛洲坝梯级水电站运行规程和算法优化后的水力数据,制定三峡-葛洲坝短期发电计划；最后,在分析了上述单目标遗传算法在处理具有高维度、非线性、寻优范围大等特点的梯级水电站优化调度问题时,存在权重系数受主观因素影响大、优化结果单一、计算效率较低等缺陷的前提下,采用了对连续决策变量寻优效果较好的多目标优化算法—非支配排序遗传算法-Ⅱ型(NSGA-Ⅱ)进行优化调度模型求解,得到优化调度问题的非劣解集。再使用基于多输入多规则模糊推理法的TOSIS决策方法对优化问题的非劣解集进行方案筛选,从而得出最符合调度需求的梯级水电站调度方案。为解决三峡-葛洲坝梯级联合优化调度问题提供了一种新思路。
[Abstract]:Large-scale cascade hydropower stations have the characteristics of complex basin topology, close connection between power and water, etc. Because of the above characteristics, the joint optimal operation of large-scale cascade hydropower stations is more complicated, and the large-scale cascade hydropower stations have the characteristics of large reservoir capacity and high water head. In addition, the uncertainty of grid load and inflow runoff of cascade hydropower stations has a great negative impact on the safety, stability and economic operation of hydropower stations. It is urgent to study the joint optimal dispatching of large-scale cascade hydropower stations, especially the short-term optimal dispatching, in order to quickly respond to the sudden change of grid load and inflow runoff of cascade hydropower stations, and to improve the efficiency of water head. Under the premise of ensuring the safety of cascade hydropower stations, the economic benefit and navigation rate are maximized. Based on the deep analysis of the shortcomings of the traditional optimization method in solving the joint optimal dispatching problem of cascade hydropower stations, such as "dimension disaster" and unsatisfactory optimization effect, the system engineering theory and modern intelligent optimization algorithm are used in this paper. The multi-objective optimization algorithm, multi-attribute decision making method and its application in cascade scheduling decision are studied. The specific contents of this paper are summarized as follows:. Based on the actual operation data and working conditions of the three Gorges Gezhou Dam, the hydraulic calculation process of the cascade hydropower station of the three Gorges Gezhou Dam is established based on the least square method, and the basic number and the hydraulic calculation flow are studied and analyzed in this paper. The mathematical model of short-term optimal dispatching of the three Gorges Gezhouba Cascade Hydropower Station is established, which creates the conditions for the use of intelligent optimization algorithm. Secondly, based on the established mathematical model of short-term optimal dispatching, The weight method is used to transform the multi-objective optimal scheduling problem into a single-objective problem. Based on the improved adaptive genetic algorithm based on Sigmoid curve, the model is solved. According to the operation rules of the three Gorges Gezhouba Cascade Hydropower Station and the optimized hydraulic data of the algorithm, the short term power generation plan of the three Gorges Gezhouba Dam is formulated. In this paper, the single objective genetic algorithm is used to deal with the cascade hydropower stations with high dimension, nonlinearity and large range of optimization. The weight coefficient is greatly affected by subjective factors, and the optimization result is single. On the premise of low computational efficiency, the optimal scheduling model is solved by using a multi-objective optimization algorithm, the non-dominated sorting genetic algorithm (NSGA- 鈪，

本文编号：1608847