【摘要】 针对电网的连锁过载,提出了一种基于聚类算法的故障关联区域划分方法。利用直流潮流法对初始故障支路和其余支路之间的内在关联因素进行了分析,给出了支路之间的关联作用矩阵;给出了一种评价支路连锁过载严重度的表达参量,并给出了反映任一初始故障支路造成其余支路功率越限的严重度矩阵以及连锁过载标识矩阵;利用该3种矩阵从初始故障对其余支路施加影响的角度以及其余支路遭受初始故障攻击的角度,给出了以聚类算法为核心的关联作用区域划分方法,该方法既考虑了支路之间的内在关联作用,也考虑了实际运行状态的影响,最后通过算例分析验证了该算法的合理性。更多还原

【Abstract】 Aiming at cascading overload phenomenon in power network,a method for portioning correlate areas based on clustering algorithm is proposed.By using DC flow algorithm,the correlate factors between the initial failed branch and the other branches in the power network is analyzed,and a factor that represents the correlation between the initial failed branch and the other branches is picked up.Then,based on that factor,a correlate matrix representing the correlation between any initial failed branch and the other branches is proposed.In order to evaluate the severity of a cascading overload event,an evaluation index is proposed,and based on the evaluation index,a matrix that represents the severity raised by any initial failed branch to other branches is proposed,and a matrix that marks the results of cascading overload is proposed.Then by using the correlate matrix,the severity matrix and the marking matrix,and from the view of subjecting cascading overload and actuating cascading overload,the algorithm for portioning correlate areas in power network under the condition of cascading overload is proposed.The algorithm can take the correlation between the initial failed branch and the other branches into account and can take the operation state of the power system into account.Because the cascading overload phenomenon is related to the correlation between the initial failed branch and the other branches,and related to the operation state of the power system,so the algorithm is a comprehensive one.Some examples show the rationality of the algorithm.更多还原