摘要
在分析负荷模型参数影响因素的基础上,基于多元线性回归法提出一种动态负荷模型参数实时选择法.首先,对一段时间内负荷的全部历史扰动实测数据进行负荷模型参数辨识,积累成模型参数数据库,并按照电压振荡幅值将其分为小扰动、一般扰动和大扰动负荷模型参数数据库;然后,根据仿真需要从相应类型的负荷模型参数数据库中用多元线性回归法搜索最匹配系统实时状况的负荷模型参数并分析其拟合精度;最后,用某城市群两个变电站的实测数据验证了所提方法的有效性和准确性.
This paper analyzes the factors influencing the load model parameters and proposes a real-time selection method of dynamic load model parameters based on the multiple linear regression.In the investigation,first,all historical load disturbance data measured in a certain duration are identified to obtain the load model parameters,and the identified parameters are used to construct a model parameter database that is further classified into three sub-databases respectively corresponding to small,common and large disturbances.Then,the parameters matching the real-time operation condition best are found from the sub-databases with correct disturbance type and the fitting accuracy is further analyzed.Finally,the effectiveness and accuracy of the proposed method are verified with the field measurement data collected from two substations in a metropolitan area in China.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2016年第2期107-116,共10页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(51377072)~~
关键词
动态负荷模型
时变性
多元线性回归
电压振荡幅度
扰动
拟合精度
dynamic load model
time-variability
multiple linear regression
voltage oscillation magnitude
dis turbance
fitting accuracy
