基于模糊聚类和最短路径的关键输电断面确定新方法

发布时间：2018-10-09 09:47

【摘要】：针对因支路故障或过载被切除之后的潮流转移关键输电断面如何正确快速确定问题,提出了基于模糊聚类和最短路径的潮流转移关键输电断面快速搜索新方法。该方法首先采用具有自适应选择最佳分类数的随机聚类方法,对所有线路功率构成进行分类,获得故障或过载支路的并行支路集构成初始输电断面,以避免由于人为给定分类数的不恰当所产生的多选及漏选问题;然后对初始输电断面中的每一支路,依次从故障或过载支路的一端开始搜索包括相应支路的到故障或过载支路另一端的最短路径,在该路径上所有与故障或过载支路潮流方向相同的支路,均为关键输电断面中的候选支路,以补充单纯由聚类方法确定的候选支路可能的漏选;候选支路再经简单的分布因子校核即可确定是否为关键输电断面中的支路。该方法实现了基于支路功率聚类关键输电断面选择方法和最短路径关键输电断面选择方法的有机融合和优势互补,提高了辨识精度。多个算例证实了文中方法的有效性。
[Abstract]:To solve the problem of how to determine the key sections of power flow transfer after branch fault or overload is removed correctly and quickly, a new method of fast searching for key transmission sections of power flow transfer based on fuzzy clustering and shortest path is proposed. In this method, a random clustering method with adaptive selection of optimal classification number is used to classify all line power components, and the parallel branch sets of fault or overload branches are obtained to form the initial transmission section. In order to avoid the problem of multi-selection and leakage caused by the improper number of artificial given classifications; and then for each branch of the initial transmission section, Starting in turn from one end of the fault or overload branch to the shortest path from the fault or overload branch to the other end of the fault or overload branch on which all branches in the same direction as the fault or overload branch flow, All of them are candidate branches in key transmission sections to supplement the possible missing selection of candidate branches determined only by clustering method, and the candidate branches can be determined by simple distribution factor checking to determine whether the candidate branch is a branch in the critical transmission section. This method realizes the organic fusion and complementary advantages of the key transmission section selection method based on branch power clustering and the shortest path key transmission section selection method, and improves the identification accuracy. Several examples show the effectiveness of the proposed method.
【作者单位】：同济大学电子与信息工程学院;天津大学电气与自动化工程学院;国网安徽省电力公司调度中心;北京科东电力控制系统有限责任公司;
【分类号】：TM73

【参考文献】