GRU neural network load modeling based on community division

doi:10.3969/j.issn.2097-0706.2022.02.003

Abstract

Abstract:

The extensive access of distributed power supplies increases the complexity of power distribution networks,resulting in the difficulty of the description by load models.In addition,modeling for the whole distribution network and each node in it is inapplicable for the ever-expanding distribution network.Therefore,a Gated Recurrent Unit （GRU） neural network load modeling method based on community division is proposed.Firstly,the GN algorithm is used to partition the topology of the distribution network.Then,load model for each community is constructed based on GRU neural network,in which degree centrality is introduced into the input feature vectors to measure the importance of nodes in the network topology.Finally,modeling analysis is carried out on a 10 kV distribution network.The results show that the proposed modelling mothed is of higher accuracy and computational efficiency.

Key words: distributed generation, load model, community division, degree centrality, GRU neural network

CLC Number:

TK01：TM74

ZHAO Shengjun, ZHANG Kaipeng, FU Xinquan, SI YingLian, LIU Zhidong, ZHOU Dengyu. GRU neural network load modeling based on community division[J]. Integrated Intelligent Energy, 2022, 44(2): 15-20.

Figures/Tables 7

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Fig.5

Table 2

References 18

[1]	鞠平, 马大强. 电力系统负荷建模 [M]. 2版.北京: 中国电力出版社, 2008.
[2]	李培强, 李欣然, 林舜江. 电力负荷建模研究述评[J]. 电力系统及其自动化学报, 2008,20(5):56-64,123.
	LI Peiqiang, LI Xinran, LIN Shunjiang. Critical review on synjournal load modeling[J]. Proceedings of the CSU-EPSA, 2008,20(5):56-64,123.
[3]	李培强, 曾小军, 李欣然, 等. 基于神经网络的分布式电源统一等效建模及其在PSASP中的应用[J]. 电网技术, 2016,40(4):1224-1229.
	LI Peiqiang, ZENG Xiaojun, LI Xinran, et al. Unified equivalent modeling of distributed generation using artificial neural network and its application in PSASP[J]. Power System Technology, 2016,40(4):1224-1229.
[4]	鞠平, 郭德正, 曹路, 等. 含主动负荷的综合电力负荷建模研究综述与展望[J]. 河海大学学报(自然科学版), 2020,48(4):367-376.
	JU Ping, GUO Dezheng, CAO Lu, et al. Review and prospect of modeling on generalized synjournal electric load containing active loads[J]. Journal of Hohai University(Natural Sciences), 2020,48(4):367-376.
[5]	赵静波, 鞠平, 施佳君, 等. 电力系统负荷建模研究综述与展望[J]. 河海大学学报(自然科学版), 2020,48(1):87-94.
	ZHAO Jingbo, JU Ping, SHI Jiajun, et al. Review and prospects for load modeling of power system[J]. Journal of Hohai University(Natural Sciences), 2020,48(1):87-94.
[6]	黄俊铭, 朱建全, 庄远灿. 基于动态RBF神经网络的广义电力负荷建模[J]. 电网技术, 2018,42(2):591-597.
	HUANG Junming, ZHU Jianquan, ZHUANG Yuancan. Generalized power load modeling based on dynamic RBF neural network[J]. Power System Technology, 2018,42(2):591-597.
[7]	LUO T Y, ZHU J Q, HUANG J M. Generalized power load modeling method with multiple RBF neural network model structures[C]//2018 International Conference on Power System Technology (POWERCON). 2018:417-423.
[8]	LI W B, XING L H, HUANG K Y, et al. Generalized load modeling method based on clustering and neural network[C]//2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). 2020:619-622.
[9]	ZENG P, LI H P, HE H B, et al. Dynamic energy management of a microgrid using approximate dynamic programming and deep recurrent neural network learning[J]. IEEE Transactions on Smart Grid, 2019,10(4):4435-4445. doi: 10.1109/TSG.5165411
[10]	CHO K, MERRIENBOER B V, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[C]//Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2014:1724-1734.
[11]	LI Y L, YANG J Y, WANG H X, et al. Dynamic equivalent modeling for microgrid based on GRU[C]//Energy Reports, 2020(6):1291-1297.
[12]	魏震波, 关翔友, 刘梁豪. 电网社区结构发现方法及其应用综述[J]. 电网技术, 2020,44(7):2600-2609.
	WEI Zhengbo, GUAN Xiangyou, LIU Lianghao. Overview of power community structure discovery algorithms and thier application in power grid analysis[J]. Power System Technology, 2020,44(7):2600-2609.
[13]	管霖, 吴亮, 卓映君, 等. 面向暂态电压控制的大电网区域划分方法[J]. 电网技术, 2018,42(11):3753-3759.
	GUAN Lin, WU Liang, ZHUO Yingjun, et al. Large power system partitioning method of transient voltage control[J]. Power System Technology, 2018,42(11):3753-3759.
[14]	周炳华, 王洋, 李峰, 等. 城市能源互联网视角下的主动配电网规划设计与策略研究[J]. 华电技术, 2021,43(1):59-65.
	ZHOU Binghua, WANG Yang, LI Feng, et al. Research on active distribution network planning and design strategy from the perspective of urban energy internet[J]. Huadian Technology, 2021,43(1):59-65.
[15]	GIRVAN M, NEWMAN M E J. Community structure in social and biological networks[J]. Proceedings of the National Academy of Sciences of the United States of America, 2002,99(12):7821-7826.
[16]	ZHENG C, WANG S R, LIU Y L, et al. A novel equivalent model of active distribution networks based on LSTM[J]. IEEE transactions on neural networks and learning systems, 2019,30(9):2611-2624. doi: 10.1109/TNNLS.5962385
[17]	姚程文, 杨苹, 刘泽健. 基于CNN-GRU混合神经网络的负荷预测方法[J]. 电网技术, 2020,44(9):3416-3424.
	YAO Chengwen, YANG Ping, LIU Zejian. Load forecasting method based on CNN-GRU hybrid neural network[J]. Power System Technology, 2020,44(9):3416-3424.
[18]	彭维珂, 聂椿明, 陈衡, 等. 基于智能算法的空冷火电机组负荷预测研究[J]. 华电技术, 2021,43(3):57-64.
	PENG Weike, NIE Chunming, CHEN Heng, et al. Study on load forecasting for air cooling thermal power units based on intelligent algorithm[J]. Huadian Technology, 2021,43(3):57-64.

GRU层数	Dense层数	收敛时间/s	均方根误差
2	1	7.71	0.068
2	2	7.62	0.059
3	2	9.20	0.063
3	3	9.13	0.057
4	3	10.82	0.066
4	4	10.91	0.063

负荷模型	训练时间/s	均方根误差
GN-GRU	7.66	0.035
All-GRU	9.13	0.057