Home > View > Focus

GCN-LSTM spatiotemporal-network-based method for post-disturbance frequency prediction of power systems

2022-03-22

【论文推荐】中国工程院院士杨奇逊等：基于 GCN-LSTM 时空网络的电力系统扰动后频率响应预测方法

文章导读

随着互联电网规模的扩大，扰动后电力系统频率响应的时空分布特征变得愈发明显，准确预测电网频率响应可以为电网安全控制提供有效支撑。然而，传统基于模型的频率响应预测方法需要建立精确的电力系统元件模型，其计算时间长，无法满足在线运行的要求，数据驱动的方法以其计算快速性和不依赖与模型的特性得到了广泛应用。因此，本文提出了一种基于图卷积神经网络（graph convolutional network，GCN）和长短期记忆网络（a long short-term memory，LSTM）组成的时空网络的频率预测模型。在所提方法中，来自同步相量测量单元测量的数据作为时空网络输入，使用嵌入电网拓扑信息的改进 GCN 提取空间维度特征以及 LSTM 提取时间维度特征。进一步，利用滚动更新的方式，训练时空网络回归模型，实现异步频率序列的预测。在 IEEE 39 节点和 IEEE 118 节点系统上验证了所提方法的有效性和抗噪性。

GCN-LSTM spatiotemporal-network-based method for post-disturbance frequency prediction of power systems

基于 GCN-LSTM 时空网络的电力系统扰动后频率响应预测方法

Dengyi Huang, Hao Liu, Tianshu Bi, Qixun Yang

(State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources,(North China Electric Power University) Changping District, Beijing 102206, P. R. China)

收听作者1分钟语音介绍

GCN-LSTM Spatiotemporal Network Based Frequency Prediction

Abstract

Owing to the expansion of the grid interconnection scale, the spatiotemporal distribution characteristics of the frequency response of power systems after the occurrence of disturbances have become increasingly important. These characteristics can provide effective support in coordinated security control. However, traditional model-based frequency- prediction methods cannot satisfactorily meet the requirements of online applications owing to the long calculation time and accurate power-system models. Therefore, this study presents a rolling frequency-prediction model based on a graph convolutional network (GCN) and a long short-term memory (LSTM) spatiotemporal network and named as STGCN-LSTM. In the proposed method, the measurement data from phasor measurement units after the occurrence of disturbances are used to construct the spatiotemporal input. An improved GCN embedded with topology information is used to extract the spatial features, while the LSTM network is used to extract the temporal features. The spatiotemporal-network-regression model is further trained, and asynchronous-frequency-sequence prediction is realized by utilizing the rolling update of measurement information. The proposed spatiotemporal-network-based prediction model can achieve accurate frequency prediction by considering the spatiotemporal distribution characteristics of the frequency response. The noise immunity and robustness of the proposed method are verified on the IEEE 39-bus and IEEE 118-bus systems.

Keywords

Synchronous phasor measurement, Frequency-response prediction, Spatiotemporal distribution characteristics, Improved graph convolutional network, Long short-term memory network, Spatiotemporal-network structure.

Fig.1 Framework of the proposed frequency-prediction method

Fig.2 Power-system structure based on PMU measurement

Fig.3 Structure of the IGCN

Fig.4 Structure of the LSTM network unit

Fig.5 Structures of the gates in the LSTM unit

Fig.6 Proposed spatiotemporal network by rolling training

Fig.7 Structure of the IEEE 10-machine 39-bus system

Fig.8 Frequency prediction results on bus 26 achieved by using different methods

Fig.9 Frequency prediction accuracy of the proposed method

Fig.10 Structure of IEEE 118-bus system

本文引文信息

Dengyi Huang, Hao Liu, Tianshu Bi, Qixun Yang (2022) GCN-LSTM spatiotemporal-network-based method for

post-disturbance frequency prediction of power systems. Global Energy Interconnection, 5(1):96-107

黄登一，刘灏，毕天姝，杨奇逊（2022）基于 GCN-LSTM 时空网络的电力系统扰动后频率响应预测方法. 全球能源互联网（英文），5(1): 96-107

Biographies

Dengyi Huang

was born in China in 1994. He received his B.Eng. degree in electrical engineering and its automation from North China Electric Power University, Beijing, China, in 2017. He is currently pursuing the Ph.D degree in electrical engineering at North China Electric Power University,

Beijing, China. His research interests include synchrophasor-based power system dynamic response prediction, propagation and control.

Hao Liu

was in Shandong, China, in 1985. He received the Ph.D. degrees in electrical engineering from North China Electric Power University in 2015. (Corresponding author).

He is currently an Associate Professor with North China Electric Power University. His research interests include synchronized measurement device, calibration, applications and disturbance propagation.

Tianshu Bi

received the Ph.D. degree in electrical and electronic engineering from the University of Hong Kong, Hong Kong, China, in 2002.

She is currently a Professor at North China Electric Power University, Beijing, China. Her research interests include power system protection and control, and synchrophasor measurement techniques and its applications.

Qixun Yang

was born in Shanghai, China, in 1937. He received the B.S. and Ph.D. degrees in Electrical Engineering from Zhejiang University, China, and South Wales University, Australia in 1960 and 1982, respectively.

He is currently a Chinese Academician of engineering. His research interests include power system protection and control, and substation automation.

编辑：王彦博

审核：王伟

郑重声明

根据国家版权局相关规定，纸媒、网站、微博、微信公众号转载、摘编本网站作品，需包含本网站名称、二维码等关键信息，并在文首注明《全球能源互联网》原创。个人请按本网站原文转发、分享。

GCN-LSTM spatiotemporal-network-based method for post-disturbance frequency prediction of power systems

【论文推荐】中国工程院院士 杨奇逊等：基于 GCN-LSTM 时空网络的电力系统扰动后频率响应预测方法

郑重声明

【论文推荐】中国工程院院士杨奇逊等：基于 GCN-LSTM 时空网络的电力系统扰动后频率响应预测方法