基于自适应分区和SFVMD-LSTM伪量测建模的新型配电系统抗差状态估计OA北大核心CSTPCD
Novel Distribution System Robust State Estimation Based on Adaptive Partitioning and SFVMD-LSTM Pseudo-Measurement Modeling
分布式资源大量接入使配电网运行机理愈加复杂,多类型不良数据、电网规模扩大等因素给新型配电系统的精准状态估计带来了新的技术挑战.提出一种基于自适应分区和时空变分模态分解-长短期记忆(spatiotemporal feature variational mode decomposition-long short term memory,SFVMD-LSTM)网络伪量测建模的新型配电系统抗差状态估计模型.在计及节点电气灵敏度的基础上,考虑不良数据分布特征,通过改进吉尔文-纽曼(Girvan and Newman,GN)分区方法克服传统GN算法在适应量测数据质量变化中的不足;利用子区域内节点多源负荷数据,提出基于SFVMD-LSTM的伪量测建模方法,提高加权最小二乘(weighted least square,WLS)估计的量测冗余度,解决传统方法状态估计精度低和抗差能力不足问题.算例仿真与结果对比分析表明,所提方法的估计精度、效率均高于传统WLS和快速解耦估计法.
The large number of accesses of distributed resources leads to more and more complex distribution network operation mechanism,as well as multiple types of undesirable data,the expansion of the grid scale and other factors bring new technical challenges to the accurate state estimation of the new distribution system.This paper proposes a new distribution system robust state estimation model based on adaptive partitioning and Spatiotemporal feature variational mode decomposition-Long Short-term Memory(SFVMD-LSTM)pseudo-measurement modeling.On the basis of taking into account the electrical sensitivity of the nodes,considering the distribution characteristics of the poor data,overcoming the shortcomings of the traditional Girvan and Newman(GN)algorithm in adapting to the changes in the quality of the measurement data by improving GN partitioning method,and utilizing the multi-source load data of the nodes in the subregion,the pseudo-measurement modeling method based on the SFVMD-LSTM is proposed,which improves the weighted least square(WLS)estimation.The pseudo-measurement modeling method based on SFVMD-LSTM is proposed to improve the measurement redundancy of WLS estimation,and to solve the problem of low accuracy and insufficient tolerance of traditional state estimation.The estimation accuracy and efficiency of the proposed method are higher than those of the traditional WLS and the Fast decoupling state estimation through example simulation and result comparison analysis.
何振武;姜飞;欧阳卫;刘利波;曾子豪;何桂雄
长沙理工大学电网防灾减灾全国重点实验室,长沙市 410114国网湖南综合能源服务有限公司,长沙市 410007中国电力科学研究院有限公司,北京市 100192
动力与电气工程
新型配电系统不良数据自适应分区时空变分模态分解状态估计
novel distribution systembad dataadaptive partitioningspatiotemporal feature variational mode decompositionstate estimation
《电力建设》 2024 (010)
78-89 / 12
This work is supported by the National Natural Science Foundation of China(No.52377166). 国家自然科学基金项目(52377166)
评论