摘要
为更有效地利用接地网腐蚀诊断结果,提出了一种借助可测性矩阵分解进行可信度分析的方法。建立了一个新的可测性矩阵,该矩阵只与接地网拓扑结构和测试信息有关,且不需进行偏导运算,并建立了接地网非线性腐蚀诊断方程,利用高斯–若当消去法从可测性矩阵中得到支路模糊集和完全可测支路集合。提出了利用模糊集中元件诊断结果的倍率分散系数进行可信度排序的方法,并给出了这些元件诊断结果可确诊的3个判断准则。对110kV蒲城变电站接地网仿真测试验证了算法的有效性。
To make the best use of results of corrosion diagnosis for grounding grids, a novel approach for creditability analysis by means of testability matrix factorization is presented. A new testability matrix was set up, which was only related to topological structure and test information without partial derivative operation, and then nonlinear diagnostic equations were established. The branch ambiguity groups and a set of surely testable branches were obtained through the testability matrix by the Gauss-Jordan elimination method. Then, by means of multiple distribution coefficients of components diagnosis results in ambiguity groups, the ordering of creditability can be carried out. Three principles were put forward, by which these components diagnosis results can be confirmed. The effectiveness of the algorithm was validated by the simulation of 1 lOkV Pucheng substation grounding grids. The creditability analysis is of great significance to corrosion detection of grounding grids by making full use of the diagnosis results
出处
《中国电机工程学报》
EI
CSCD
北大核心
2010年第7期98-103,共6页
Proceedings of the CSEE
关键词
接地网
腐蚀诊断
可测性矩阵
可信度
grounding grid
corrosion diagnosis
testability matrix
creditability