期刊 基于自适应核密度的含PET交直流混合配电网状态估计  

Adaptive Kernel Density Theory Based State Estimation for Hybrid AC/DC Distribution Network with PET

作  者:贾宏杰[1] 司佳 穆云飞[1] 刘阳升 

Jia Hongjie;Si Jia;Mu Yunfei;Liu Yangsheng(Key Laboratory of Smart Grid of Ministry of Education(Tianjin University),Tianjin 300072,China;School of Mechanical and Electrical Engineering,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区:[1]智能电网教育部重点实验室(天津大学),天津300072 [2]桂林电子科技大学机电工程学院,桂林541004

出  处:《天津大学学报:自然科学与工程技术版》2021年第7期754-762,共9页Journal of Tianjin University(Science and Technology)

Adaptive Kernel Density Theory Based State Estimation for Hybrid AC/DC Distribution Network with PET

基  金:国家重点研发计划资助项目(2017YFB0903300);国家自然科学基金资助项目(52061635103).

摘  要:围绕如何提高含电力电子变压器(PET)的交直流混合配电网状态估计的精度问题,本文考虑PET不同控制方式下的端口量测特性,提出了基于自适应核密度的交直流混合配电网状态估计新方法,并分析了PET不同控制方式下局部量测配置对状态估计精度的影响规律.首先考虑量测中可能存在的坏数据以及交流部分网络的三相不对称特性,构建了基于自适应核密度的三相状态估计模型;进而提出了考虑三相不对称交流网络、直流网络以及PET量测的带宽确定方法,可在状态估计迭代过程中根据残差大小自动生成近似最优带宽、初始带宽及修正带宽,使得坏数据在迭代过程中逐渐降权,以提升坏数据辨识精度,并通过量测标准差约束克服好数据被误辨识而导致的局部不可观测问题;同时定性地分析了PET不同控制方式下不同类型量测对状态估计精度的影响特点,给出了PET量测配置建议,以进一步改善状态估计精度.最后通过构建含PET的22个节点交直流混合配电网算例验证表明,本文自适应核密度状态估计方法的精度相较于传统加权最小二乘法获得明显提升,电压幅值和相角估计精度分别提高50%和35%.且PET电压量测精度对状态估计结果有显著影响,当其精度增加1倍时,电压幅值和相角精度分别提高30%和10%.

To improve the accuracy of state estimation for a hybrid AC/DC distribution network with power electronic transformer(PET),a new state estimation method based on adaptive kernel density is proposed,considering the measurement characteristics of PET ports under different control modes.In addition,the influence of local measurement configuration on state estimation accuracy is analyzed under different control modes of PET.First,considering the possible bad data in the measurement and the three-phase asymmetric characteristics of the AC part of the network,a three-phase state estimation model based on adaptive kernel density is constructed.Next,a bandwidth determination method is proposed considering the three-phase asymmetric AC network,DC network,and PET measurement.This method can automatically generate the approximate optimal bandwidth,initial bandwidth,and modified bandwidth according to the residual value in the iterative process of state estimation to improve the identifycation accuracy of bad data,the equivalent weights of which can be autumnally descended.The problem of erroneous identification resulted in partial unobservability,which is overcome by measuring the standard deviation constraint.At the same time,the influence of different PET control modes on the accuracy of state estimation is analyzed qualitatively,and the PET measurement configuration scheme is suggested,which further improved the estimation accuracy.Finally,a 22-bus AC/DC hybrid distribution network with PET is constructed to verify the accuracy of the proposed state estimation method compared with the traditional weighted least square method;the accuracy of the voltage magnitude and the phase angle improved by 50%and 35%,respectively.The accuracy of PET measurement has a significant impact on the state estimation results,the accuracy of the voltage magnitude and the phase angle increased by 30%and 10%,respectively,also doubling the accuracy of the voltage magnitude in PET.

关 键 词:电力电子变压器 配电网 核密度估计 状态估计 

power electronic transformer(PET) distribution network kernel density estimation state estimation 

分 类 号:TM76[电气工程—电力系统及自动化]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关文献:

正在载入数据...

北京电子科技职业学院特色库 版权所有 ©2018