基于聚类机群关联拓扑的时空图记忆风速超短期预测  

Ultra-short-term Wind Speed Forecasting Based on Spatio-temporal Graph Memory of Clustering Wind Turbine Group Association Topology

在线阅读下载全文

作  者:潘超[1] 蒋迪遥 李宝聚 孙勇 郝成亮 PAN Chao;JIANG Diyao;LI Baoju;SUN Yong;HAO Chengliang(Key Laboratory of Modern Power System Simulation and Control&Renewable Energy Technology(Northeast Electric Power University),Ministry of Education,Jilin 132012,Jilin Province,China;State Grid Jilin Electric Power Co.,Ltd.,Changchun 130031,Jilin Province,China)

机构地区:[1]现代电力系统仿真控制与绿色电能新技术教育部重点实验室(东北电力大学),吉林省吉林市132012 [2]国网吉林省电力有限公司,吉林省长春市130031

出  处:《电网技术》2023年第11期4607-4618,共12页Power System Technology

基  金:国家重点研发计划项目(2022YFB2404001)。

摘  要:为提高规模化风电场风速的预测精度及计算效率,提出一种聚类机群最优关联拓扑,并构建时空图记忆模型预测风速。分析机群平均风速波动特性,构建平抑度指标;考虑风机间风向关联特性,定义风向相似度因子,并嵌入k-means聚类,提高类内风速互补性。结合互信息量化分析各子机群风机的相关性,构建最优有向关联拓扑。结合关联拓扑及风速关联属性建立机群风速时空图数据集合,输入时空图记忆网络,利用图注意力提取空间特征,并结合记忆网络处理时序信息,输出机群平均风速超短期预测结果。最后将模型应用于实际风电场的风速预测,通过对比分析,验证了所提方法的准确性和有效性。In order to improve the prediction accuracy and efficiency of the wind speed in large-scale wind farms,an optimal correlation topology of a clustering wind turbine group is proposed,and a spatio-temporal graph memory model is constructed to predict the wind speed.The average wind speed fluctuation characteristics of the cluster are analyzed,and the stability index is structured.Based on this,the k-means clustering is embedded to improve the intra-class wind speed complementarity.Combined with the mutual information quantitative analysis of the correlation between the sub-cluster fans,the optimal directed correlation topology is built.Based on the correlation topology and the wind speed correlation attributes,the spatial-temporal graph data set of the cluster wind speed is established by inputting the spatial-temporal graph memory network.The spatial features are extracted by the graph attention,and the time series information is processed by the memory network to output the ultra-short-term prediction results of the cluster average wind speed.Finally,the model is applied to the actual wind farm wind speed prediction,and the accuracy and effectiveness of the proposed method are verified through comparative analysis.

关 键 词:机群平均风速 改进k-means聚类 最优有向关联拓扑 时空图记忆网络 

分 类 号:TM614[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...

相关的主题
相关的作者对象
相关的机构对象