High fault-tolerant distribution network state estimation method based on gated graph neural network

doi:10.3969/j.issn.2097-0706.2023.06.001

Abstract

Abstract:

With the increase of renewable energy's penetration rate in power systems， it is essential to make real-time， accurate and highly fault-tolerant state estimations for distribution networks to cope with the intermittency of renewable energy and keep safe operation of the power grid. Since the assembly level of distribution network measurement devices is incomplete and the model-driven state estimation can hardly adapt to the high uncertainty of environment， the fused measurement data from SCADA/PMU is adopted to train the gated graph neural network （GGNN）.Then， a high fault-tolerant distribution network state estimation method based on GGNN is proposed. It can obtain the spatio-temporal relationship between the measurement and the state estimation by using the graph convolutional layer and GRU-like to extract high-dimensional spatio-temporal features of the measurement. The proposed algorithmic solution is assessed and validated based on an IEEE 33-bus system and an IEEE 118-bus system， respectively. The assessment result show that GGNN can effectively fit the space-time mapping of measurement and state data with a higher accuracy and robustness compared with the traditional Weighted Least Squares （WLS） and Multi-layer Perceptron （MLP）.

Key words: renewable energy, distribution network, state estimation, high fault tolerance, fused measurement, gated graph neural network, least squares method, multi-layer perceptron

CLC Number:

TK01：TM732

LIU Yixian, WANG Yubin, YANG Qiang. High fault-tolerant distribution network state estimation method based on gated graph neural network[J]. Integrated Intelligent Energy, 2023, 45(6): 1-8.

Figures/Tables 17

Fig.1

Table 1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Table 2

Table 3

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Table 4

Table 5

Training set Loss and test set δMSE under different Batch size

项目	Batch
项目	32	60	96	128
训练集Loss( $× 10 - 5$ )	3.76	1.47	2.34	3.67
测试集δ_MSE	0.067	0.064	0.082	0.079

Table 5

Table 6

Training set Loss and test set δMSE under different Eepoch size

项目	Epoch
项目	50	100	150	200
训练集Loss( $× 10 - 5$ )	7.21	1.01	5.49	4.62
测试集δ_MSE	0.076	0.068	0.068	0.069

Table 6

References 24

[1]	马钊, 周孝信, 尚宇炜, 等. 未来配电系统形态及发展趋势[J]. 中国电机工程学报, 2015, 35(6): 1289-1298.
	MA Zhao, ZHOU Xiaoxin, SHANG Yuwei, et al. Form and development trend of future distribution system[J]. Proceedings of the CSEE, 2015, 35(6): 1289-1298.
[2]	MELIOPOULOS A P S, POLYMENEAS E, TAN Z, et al. Advanced distribution management system[J]. IEEE Trans on Smart Grid, 2013, 4(4): 2109-2117. doi: 10.1109/TSG.2013.2261564
[3]	褚晨杰, 吕干云, 吕经纬, 等. 基于PMU/SCADA混合量测的电力系统鲁棒状态估计[J]. 南京工程学院学报(自然科学版), 2020, 18(4):14-19.
	CHU Chenjie, LYU Ganyun, LYU Jingwei, et al. Robust state estimation of power system based on PMU and SCADA hybrid measurement[J]. Journal of Nanjing Institute of Technology (Natural Science Edition), 2020, 18(4):14-19.
[4]	丁宏恩, 戴则梅, 霍雪松, 等. 基于混合量测的二次线性状态估计方法及其工程应用[J]. 电力系统保护与控制, 2016, 44(13):163-170.
	DING Hong'en, DAI Zemei, HUO Xuesong, et al. Hybrid measurement based secondary linear state estimation method and its engineering application[J]. Power System Protection and Control, 2016, 44(13):163-170.
[5]	FILHO M B D C, SOUZA J C S D. Forecasting-aided state estimation—Part I: Panorama[J]. IEEE Transactions on Power Systems, 2009, 24(4): 1667 -1677. doi: 10.1109/TPWRS.2009.2030295
[6]	丁军策, 蔡泽祥, 王克英. 基于广域测量系统的混合量测状态估计算法[J]. 中国电机工程学报, 2006, 26(2):58-63.
	DING Junce, CAI Zexiang, WANG Keying. Mixed measurements state estimation based on WAMS[J]. Proceedings of the CSEE, 2006, 26(2):58-63.
[7]	赵红嘎, 薛禹胜, 汪德星, 等. 计及PMU支路电流相量的状态估计模型[J]. 电力系统自动化, 2004, 28(17):37-40.
	ZHAO Hongga, XUE Yusheng, WANG Dexing, et al. Station estimation model with PMU current phasor measurement[J]. Automation of Electric Power Systems, 2004, 28(17): 37-40.
[8]	于尔铿. 电力系统状态估计[M]. 北京: 水利电力出版社,1985.
[9]	王玉彬. 基于PMU/SCADA量测数据的配电网状态估计方法研究[D]. 北京: 北京交通大学, 2019.
	WANG Yubin. Research on distribution network state estimation method based on PMU/SCADA measurement data[D]. Beijing: Beijing Jiaotong University, 2019.
[10]	李炜, 严川, 盛庆博, 等. 大数据背景下智能配电网运营管理方法研究[J]. 华电技术, 2021, 43(8): 33-40.
	LI Wei, YAN Chuan, SHENG Qingbo, et al. Study on smart distribution network operation and management methods in the context of big data[J]. Huadian Technology, 2021, 43(8):33-40.
[11]	罗玉春, 王毅. PAS网络建模流程的设计[J]. 电力工程技术, 2015, 34(5):40-42.
	LUO Yuchun, WANG Yi. Design of power grid modeling process in advanced application software package[J]. Electric Power Engineering Technology, 2015, 34(5):40-42.
[12]	周寅飞. 基于内点法和离散粒子群算法的输电网参数辨识[J]. 电力工程技术, 2014, 33(1):22-25.
	ZHOU Yinfei. Hybrid strategy based on interior point method and DPSO for electricity transmission network parameter estimation[J]. Electric Power Engineering Technology, 2014, 33(1):22-25.
[13]	王义, 杨志伟, 吴坡, 等. 计及高比例分布式光伏能源接入的配电网状态估计[J]. 综合智慧能源, 2022, 44(10): 12-18. doi: 10.3969/j.issn.2097-0706.2022.10.002
	WANG Yi, YANG Zhiwei, WU Po, et al. State estimation for the distribution network with high-proportion distributed photovoltaic energy[J]. Integrated Intelligent Energy, 2022, 44(10): 12-18. doi: 10.3969/j.issn.2097-0706.2022.10.002
[14]	谢铁明, 卫志农, 袁安建. 基于超短期负荷预测的电力系统动态状态估计[J]. 电力工程技术, 2009, 28(1):8-10.
	XIE Tieming, WEI Zhinong, YUAN Anjian. Dynamic state estimation based on ultra short-term prediction of load[J]. Electric Power Engineering Technology, 2009, 28(1):8-10.
[15]	孙国强, 钱嫱, 陈亮, 等. 基于深度信念网络伪量测建模的配电网状态估计[J]. 电力自动化设备, 2018, 32(12): 94-99.
	SUN Guoqiang, QIAN Qiang, CHEN Liang, et al. State estimation of distribution system based on pesudo measurement modeling using deep belief network[J]. Electric Power Automation Equipment, 2018, 32(12):94-99.
[16]	赵巍岳, 靳松, 吕天成. 基于神经网络的状态估计方法研究[J]. 电力系统保护与控制, 2018, 46(22):109-115.
	ZHAO Weiyue, JIN Song, LYU Tiancheng. Research on state estimation based on artificial neural network[J]. Power System Protection and Control, 2018, 46(22): 109-115.
[17]	周念成, 廖建权, 王强钢, 等. 深度学习在智能电网中的应用现状分析与展望[J]. 电力系统自动化, 2019, 43(4):180-191.
	ZHOU Niancheng, LIAO Jianquan, WANG Qianggang, et al. Analysis and prospect of deep learning application in smart grid[J]. Automation of Electric Power Systems, 2019, 43(4):180-191.
[18]	DUBEY A, CHAKRABARTI S, TERZIJA V. SCADA and PMU measurement based methods for robust hybrid state estimation[J]. Electric Power Components and Systems, 2019, 47(9-10): 849-860. doi: 10.1080/15325008.2019.1627606
[19]	SKOK S, IVANKOVIC I, CERINA Z. Hybrid state estimation model based on PMU and SCADA measurements[J]. IFAC-PapersOnLine, 2016, 49(27):390-394.
[20]	BRONSTEIN M M, BRUNA J, LECUN Y, et al. Geometric deep learning: Going beyond euclidean data[J]. IEEE Signal Processing Magazine, 2017, 34(4):18-42.
[21]	LI Y J, DANIEL T, MARC B, et al. Gated graph sequence neural networks[C]// 4th International Conference on Learning Representations.San Juan, 2016:1884-2021.
[22]	ZIMMERMAN R D, MURLLO-SANCHEZ C E, THOMAS R J. MATPOWER: Steady-state operations,planning, and analysis tools for power systems research and education[J]. IEEE Transactions on Power Systems, 2011, 26(1):12-19. doi: 10.1109/TPWRS.2010.2051168
[23]	NYISO. Load data[EB/OL].(2021-02-13)[2022-10-22]. http://www.nyiso.com/.
[24]	黄蔓云, 孙国强, 卫志农, 等. 基于历史数据挖掘的配电网态势感知方法[J]. 电网技术, 2017, 41(4):1139-1145.
	HUANG Manyun, SUN Guoqiang, WEI Zhinong, et al. A novel situation awareness approach based on historical data-mining model in distribution networks[J]. Power System Technology, 2017, 41(4):1139-1145.

项目	时刻1	时刻2	时刻3	时刻4	时刻5
预测值（标幺值）	0.988 7	1.003 0	0.990 3	0.997 8	0.994 3
真实值（标幺值）	0.982 6	1.007 9	0.981 4	1.025 8	0.982 5
误差/%	0.620	0.486	0.907	2.730	1.201
准确率/%	99.380	99.514	99.093	97.280	98.799

节点系统	PMU节点编号
IEEE 33	3,6,9,11,14,17,19,22,24,26,29,32
IEEE 118	1,6,8,12,15,19,21,27,28,32,34,40,45,49,53,56,62,65,72,75,77,80,85,86,90,92,94,102,105,110,116

数据集	噪声
历史潮流真值	—
历史融合量测	SCADA系统量测误差标准值为0.020，均值为0；PMU幅值量测误差标准差为0.005；PMU相角量测误差标准差为0.002，均值为0
高噪声下融合量测	伪量测额外添加20%高斯白噪声

方法	电压幅值相对误差	电压相角相对误差
WLS	7.02	7.59
MLP	4.11	6.58
GGNN	2.89	5.11