摘要
Givens正交变换法是提高电力系统状态估计数值稳定性的有效方法,但是在变换过程中出现的大量非零注入元影响了其计算效率。按最小度原则的列编号排序和VPAIR逐列变转轴消元策略可以有效减少注入元数目。在上述消元策略的基础上,提出在消元过程中动态调整量测排序的改进策略。所提策略在选择合适旋转对进行Givens变换的同时,进行量测排序的动态优化,可以进一步减少注入元数目。将改进策略与VPAIR策略在不同规模仿真系统的仿真结果进行比较,表明所提改进策略具有更高的计算效率,能够更好地满足在线状态估计的要求。
Givens orthogonal transformation method is an effective way to improve the numerical stability of power system state estimation,while its computational efficiency is affected by a large number of non-zero fill-in elements in the transformation progress. The quantity of fill-in elements can be reduced effectively by minimum degree algorithm-based column ordering and VPAIR column oriented variable pivot strategy. Based on the above elimination strategy,an improved strategy is proposed to adjust measurement ranking dynamically in the elimination process. The proposed strategy dynamically optimizes the measurement order in the process of selecting appropriate pair for rotation,which further reduces the fill-in elements. Simulative results of the improved strategy and VPAIR strategy in different scale simulation systems are compared,verifying that the improved strategy has higher computational efficiency and can better meet the requirements of online state estimation.
出处
《电力自动化设备》
EI
CSCD
北大核心
2017年第10期166-170,共5页
Electric Power Automation Equipment
关键词
状态估计
正交变换
Givens旋转
消元策略
量测动态排序
state estimation
orthogonal transformation
Givens rotation
elimination strategy
dynamicmeasurement ranking
