利用本征正交分解进行电力系统模型降阶  

Order-reduction of Power System Model Using Proper Orthogonal Decomposition

作　　者：罗滇生[1] 李芳[1] 崔天翔 李帅虎[1] 黄根[1] LUO Diansheng;LI Fang;CUI Tianxiang;LI Shuaihu;HUANG Gen(College of Electrical and Information Engineering,Hunan University,Changsha 410082,China;Xi’an Railway Survey and Design Institute Co.,Ltd,Xi’an 710054,China)

机构地区：[1]湖南大学电气与信息工程学院,长沙410082 [2]中铁西安勘察设计研究院有限责任公司,西安710054

出　　处：《电力系统及其自动化学报》2018年第4期121-125,143,共6页Proceedings of the CSU-EPSA

摘　　要：针对电力系统模型降阶问题,提出一种基于本征正交分解POD(proper orthogonal decomposition)理论的降阶方法。首先依据动态电力系统模型建立系统样本向量集,并利用样本向量集按降序排列的奇异值以及相应的奇异值向量构造一个正交矩阵;然后着选择一个可以保持样本集基本特征信息的参数,构造POD基向量矩阵;最后利用基向量矩阵得到降阶模型。仿真计算结果表明,POD降阶模型可以保留原系统稳定性以及其输入输出动态行为,验证了所提方法的实用性。Considering the order-reduction problem of power system model,an order-reduction method based on proper orthogonal decomposition(POD)is proposed in this paper.First,a systematic sample vector set is established based on a dynamic power system model,and an orthogonal matrixis constructed by using the singular value in a descending order and the corresponding singular value vectors based on the sample vector set.Then,a POD base vector matrixis constructed by selecting a parameter that can keep the basic characteristics of the sample set.Finally,based on the base vector matrix,the order-reductionmodel of power system is obtained.The simulation calculation results show that the POD order-reduction model can keep the stability of the original system and its dynamic behavior of input/output,which verifiesthe practicability of the proposed method.

关 键 词：电力系统模型 本征正交分解(POD) 基向量 矩阵 降阶 

分 类 号：TM711[电气工程—电力系统及自动化]