摘要
对于广域阻尼控制器的实际工程应用,提出一种实用型电力系统状态空间模型辨识方法。该方法采用耦合最小二乘算法,充分利用辨识过程中的中间数据,避免重复对高维协方差矩阵迭代计算而导致的冗余过程。为进一步提高辨识精度,结合多新息辨识思想,推导多新息耦合最小二乘算法。利用系统中存在的类噪声信号,完成无附加激励下的电力系统辨识。4机2区以及10机39节点系统验证了所提方法具有辨识方便易行、计算量小、辨识精度较高的优点,对于电网结构的不断变化能方便地进行数据的在线更新。
A practical identification method of power system state space model is proposed for the engineering application of wide area damping controller,which adopts the coupling least square algorithm to make full use of the intermediate data during identification and avoid the redundancy due to the repeated iterative calculation of high dimensional eovariance identification, the multi-innovation coupling least square matrix. Combined with the multi-innovation algorithm is deduced to further improve the identification accuracy. Without any additional excitation,the noise-like signals existing in the system are used to complete the power system identification. Simulations for 4-machine 2-area and lO-machine 39-bus test systems are carried out to verify the advantages of the proposed method：simple identification process, small calculation load,high identification accuracy,and easy online data updating according to system structure change.
出处
《电力自动化设备》
EI
CSCD
北大核心
2015年第7期65-73,共9页
Electric Power Automation Equipment
基金
高等学校博士学科点专项科研基金资助项目(2011-0141110032)
西安交通大学电力设备电气绝缘国家重点实验室资助项目(EIPE13205)~~
关键词
多新息耦合最小二乘算法
电力系统
状态空间
系统辨识
广域阻尼控制器
低频振荡
multi-innovation coupling least square algorithm
electric power systems
state space
system identification
wide area damping controller
low frequency oscillation
