| 引用本文: | 刘可真,谭化平,杨春昊,李鹏,陈艳霞,普伟,赵贞焰.考虑系统分区与节点频率动态响应的电力系统区域惯量估计方法[J].电力自动化设备,2025,45(5):200-208. |
| LIU Kezhen,TAN Huaping,YANG Chunhao,LI Peng,CHEN Yanxia,PU Wei,ZHAO Zhenyan.Power system regional inertia estimation method considering system partitioning and dynamic response of node frequency[J].Electric Power Automation Equipment,2025,45(5):200-208. |
|
| 摘要: |
| 大量新能源接入电力系统将导致系统空间惯量分布差异明显并造成系统惯量下降，影响系统稳定运行。对此，提出一种考虑系统分区与节点频率动态响应的电力系统区域惯量辨识估计方法。采用改进谱聚类算法以系统导纳矩阵为输入对系统进行静态区域划分，以区域轮廓系数为分区参考指标；采用实数序列编辑距离算法动态选取区域频率最佳拟合点；采用鲍可斯-詹金斯辨识模型，赤池信息准则确定模型阶数对区域惯量进行系统参数辨识，以最小化输出误差为准则对辨识参数进行迭代优化。IEEE 10机39节点新英格兰模型验证所提方法能在误差允许范围内估计各区域惯量。 |
| 关键词: 区域惯量估计 实数序列编辑距离 谱聚类 鲍可斯-詹金斯辨识模型 赤池信息准则 |
| DOI：10.16081/j.epae.202412005 |
| 分类号:TM73 |
| 基金项目:云南电网有限责任公司科技项目(YNKJXM20180736)；云南省决策研究咨询课题(20245304) |
|
| Power system regional inertia estimation method considering system partitioning and dynamic response of node frequency |
|
LIU Kezhen1, TAN Huaping1, YANG Chunhao2, LI Peng3, CHEN Yanxia4, PU Wei1, ZHAO Zhenyan1
|
|
1.Faculty of Electric Power Engineering, Kunming University of Science and Technology, Kunming 650500, China;2.Yunnan Power Grid Co.,Ltd.,Kunming 650034, China;3.Key Laboratory of Distributed Energy Storage and Microgrid of Hebei Province, North China Electric Power University, Baoding 071003, China;4.Electric Power Research Institute, State Grid Beijing Electric Power Company, Beijing 100075, China
|
| Abstract: |
| A large number of new energy sources are connected to the power system, which will lead to a significant difference in the distribution of system spatial inertia and decrease the inertia of the system, affecting the stable operation of the system. In this regard, an identification and estimation method of regional inertia of power system considering the system partitioning and dynamic response of node frequency is proposed. An improved spectral clustering algorithm with the system conductivity matrix as input is adopted to divide the system into static regions, and the regional silhouette coefficients are taken as the partition reference indexes. An edit distance on real sequence algorithm is adopted to dynamically select the best-fitting point for the regional frequency. The Box-Jenkins model is adopted to identify the model, the Akaike information criterion is adopted to determine the model order for system parameter identification of the regional inertia, the minimized output error is used as the criteria to iteratively optimized the identification parameters. The IEEE 10-machine 39-bus New England model verifies that the proposed method can estimate the regional inertia within the error tolerance. |
| Key words: estimation of regional inertia edit distance on real sequence spectral clustering Box-Jenkins model Akaike information criterion |
