考虑风电与负荷不确定性的配电网重构研究

发布时间：2018-06-24 09:26

本文选题：配电网重构 + 风电　；参考：《东北电力大学》2017年硕士论文

【摘要】：配电网重构是配电网优化运行的一项重要措施,配电网中包含少量的联络开关和大量的分段开关,配电网重构就是重新组合配电网中这些开关的开合状态,从而使配电网运行于最佳状态。近年来,风电等分布式电源在电网中的渗透率逐年增加,配电网需要考虑随之而来的不确定性。传统的配电网重构方案已不能适应如今的配电网形势,因此有必要研究计及分布式电源影响的配电网重构。本文选取风电这一典型的分布式电源作为研究对象,建立了基于网损和网损期望值的多场景配电网重构数学模型,采用了基于“排同存异”的环路编码策略,制定了基于场景分析法和二进制粒子群算法的最优方案确定策略。基于风速的Weibull分布模型,分别研究了不考虑负荷季节变化特性的配电网重构和考虑负荷季节变化特性的配电网重构,研究中采用了基于wasserstein距离的场景划分方法对风电场景进行划分,考虑负荷季节变化特性时,采用了基于K-means的聚类方法对各季节的典型日负荷曲线进行聚类,提取出各季节的典型负荷状态,并对负荷聚类的有效性进行了验证。考虑到两参数Weibull分布模型不包含时序信息,也无法体现出风电的季节性变化特征,本文采用一种基于K-means聚类的风速场景纵横消减算法。该方法无需知道风速的概率分布函数,通过对各季节的历史风速数据纵向和横向的消减,生成各季节的风速场景集,据此研究了考虑风电季节变化特性的配电网重构,并对风电场景划分的有效性进行了验证。借助IEEE33节点配电网系统对不同情况下的配电网重构进行仿真计算,得到了各个情况下的重构方案,验证了本文建立的多场景配电网重构模型和求解方法的有效性。本文的数学模型、求解方法以及重构结果能为运行人员提供参考。
[Abstract]:Distribution network reconfiguration is an important measure in the optimal operation of distribution network. The distribution network contains a small number of contact switches and a large number of piecewise switches. Distribution network reconfiguration is to recombine the switching state of these switches in the distribution network. Thus, the distribution network runs in the best condition. In recent years, the permeability of distributed generation, such as wind power, has been increasing year by year. The traditional reconfiguration scheme of distribution network can not adapt to the current situation of distribution network, so it is necessary to study the reconfiguration of distribution network considering the influence of distributed generation. This paper selects wind power as a typical distributed power source as the research object, establishes the mathematical model of multi-scene distribution network reconfiguration based on network loss and network loss expectation value, and adopts the loop coding strategy based on "same arrangement, same storage and difference". An optimal scheme determination strategy based on scenario analysis and binary particle swarm optimization (BPSO) is proposed. Based on the Weibull distribution model of wind speed, the reconfiguration of distribution network without considering the seasonal variation of load and the reconfiguration of distribution network with consideration of seasonal variation of load are studied respectively. In the research, the scene partition method based on wasserstein distance is used to divide the wind power scene. When considering the seasonal variation of load, K-means clustering method is used to cluster the typical daily load curve of each season. The typical load states of each season were extracted and the validity of load clustering was verified. Considering that the two-parameter Weibull distribution model does not contain time series information and can not reflect the seasonal variation of wind power, this paper uses a K-means clustering algorithm to reduce wind speed scene. This method does not need to know the probability distribution function of wind speed. By reducing the historical wind speed data of each season vertically and horizontally, the scene set of wind speed in each season is generated. Based on this, the distribution network reconstruction considering the seasonal variation characteristics of wind power is studied. The validity of wind power scene partition is verified. With the help of IEEE33 node distribution network system, the reconfiguration of distribution network under different conditions is simulated and calculated, and the reconfiguration scheme in each case is obtained, which verifies the validity of the multi-scene distribution network reconfiguration model and the solution method established in this paper. The mathematical model, solution method and reconstruction result can provide reference for operators.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711

【参考文献】