摘要
首次将隐式重启动Arnoldi算法应用于大型电力系统低频机电振荡的特征值计算。证明了对重启动Arnoldi算法、广义Cayley变换与平移 逆变换等价的条件。通过算例和其它重启动Arnoldi算法的详细比较 ,表明隐式重启动Arnoldi算法收敛迅速、计算可靠且性能稳定 ,能有效计算大型电力系统中存在的特征值簇。
In this paper the implicitly restarted Arnoldi algorithm is successfully applied to large power systems for the computation of low frequency electo mechanical oscillation modes for the first time. It is proved that shift invert transformation and generalized Cayley transformation can be equivalent under certain conditions for restarted Arnoldi methods. Comparisons are made with other restarted Arnoldi methods with numerical experiments. Results show that implicitly restarted Arnoldi algorithm is fast, robust and reliable. It can compute eigenvalue clusters that exist in large power systems efficiently.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2000年第9期40-54,共15页
Proceedings of the CSEE
基金
上海市教委重点学科项目! 电力系统计算用软件与网络技术
国家重大科技攻关项目!子课题 97 31 2 1 1 1 1e"三峡发电机励磁系统研究
关键词
大型电力系统
低频机电振荡
计算方法
large power systems
low frequency electro mechanical oscillations
eigenvalue analysis method
implicitly restarted Arnoldi algorithm
spectral transformation
