基于随机矩阵谱偏离度的微网状态感知方法研究

A Method for Microgrid Situation Awareness Based on the Spectral Deviation of Random Matrix

针对大规模微网的状态监测问题,在实现快速检测及考虑节点数据控制的基础上,提出一种基于随机矩阵谱偏离度的微网状态感知方法。首先通过PBH秩判据分析微网邻接矩阵并且从中选择出感知微网数据的测控节点。然后通过所选择节点的电压及其功角序列构造的标准测控协方差矩阵,提高节点数据变化的灵敏度,并且将微网状态判断转化为分析随机矩阵谱偏离度变化程度。在得知微网状态改变的前提下,根据谱分布对应的最大特征向量元素改变量得到波动邻近测控节点,降低数据变化检测时间,满足在线快速检测的要求。最后以改进的IEEE 37节点微网为例进行仿真,仿真结果验证了所提方法的有效性。

To solve the large scale microgrid situation awareness problem, the spectral deviation of random matrix method was proposed with considering rapid detection and nodes data control. Firstly, the adjacency matrix of microgrid was analyzed by PBH rank criterion, and nodes containing measure and control were selected. Then, a covariance matrix including the voltage magnitude and the power angle sequence of selected nodes was built that could improve the sensitivity of node data change. And the status judgement of microgrid was converted to the degree analysis of the spectral deviation of random matrix. Moreover, the changing of maximum eigenvector of spectral distribution obtains the node containing measure and control near the fluctuation on the basis of microgrid situation change, which can improve the speed of detection and meet the requirement of rapid online detection. Finally, the effective of the proposed method was verified through the modified IEEE-37 bus microgrid.