基于长短期记忆网络的电力系统量测缺失数据恢复方法

doi:10.12204/j.issn.1000-7229.2021.05.001

摘要/Abstract

摘要：

随着电力系统规模不断增大,电力系统量测数据呈现快速增长趋势。然而海量数据的采集、测量、传输和存储等过程均可能出现数据缺失问题,从而威胁电网安全。针对电力系统量测缺失数据问题,文章提出了一种基于长短期记忆(long short-term memory,LSTM)网络的缺失数据恢复方法。首先,基于LSTM网络具有提取电力系统量测数据时序规律的特性,提出一种双层全连接LSTM网络模型,利用已知数据建立对缺失数据的映射。其次,为提高系统不同数据状态下的恢复精度,提出了一种随机森林状态辨识方法和考虑缺失数据位置的恢复策略。最后,利用仿真数据和实测数据验证该方法的有效性和准确性,结果表明该方法无需系统拓扑参数即可显著提高电力系统量测数据质量。

关键词: 电力系统, 量测缺失数据恢复, 长短期记忆网络, 随机森林

Abstract:

As the scale of power systems continues to increase, measurement data of power system grows rapidly. However, the process of massive data collection, measurement, transmission, and storage in power systems may lose data, which seriously threatens the safety of the power grid. Aiming at the problem of missing data in power systems, this paper proposes a missing data recovery method based on long short-term memory (LSTM) networks. Firstly, since the LSTM network can extract the characteristics of the timing law of measurement data, a double-layer and full connection LSTM network architecture is presented, which uses known data to map missing data. Furthermore, in order to improve the recovery accuracy of missing data in different states, a state identification method based on random forest and a recovery strategy considering the location of the missing data are developed. Finally, case studies by using simulation data and measurement data verify the effectiveness and accuracy of the method proposed in this paper, and the proposed method can significantly improve the data quality of the power systems without system topology modeling.

Key words: power system, recovery of missing measurement data, long short-term memory network, random forest

中图分类号:

TM73

王子馨, 胡俊杰, 刘宝柱. 基于长短期记忆网络的电力系统量测缺失数据恢复方法[J]. 电力建设, 2021, 42(5): 1-8.

WANG Zixin, HU Junjie, LIU Baozhu. Recovery Method for Missing Measurement Data of Power Systems Based on Long Short-Term Memory Networks[J]. ELECTRIC POWER CONSTRUCTION, 2021, 42(5): 1-8.

图/表 15

图1 动态数据示意图

Fig.1 Schematic diagram of dynamic data

图2 LSTM网络单元结构

Fig.2 Unit structure of LSTM network

图3 双层LSTM网络模型

Fig.3 Double-layer LSTM network model

图4 稳态数据和动态数据

Fig.4 Data in steady state and dynamic state

表1 不同位置下恢复策略

Table 1 Recovery strategy for different locations

图5 缺失数据恢复流程

Fig.5 Process of missing data recovery

图6 IEEE 10机39节点拓扑

Fig.6 Topology of IEEE 39-node system

图7 稳态和动态仿真数据

Fig.7 Simulation data in steady and dynamic state

图8 仿真数据状态辨识结果

Fig.8 State identification results of simulation data

表2 不同缺失位置对数据恢复精度的影响

Table 2 Effect of different missing positions on data recovery accuracy

表3 不同缺失占比对数据恢复精度的影响

Table 3 Effect of different missing ratios on data recovery accuracy

图9 动态缺失数据恢复结果

Fig.9 Recovery results of dynamic missing data

图10 现场实测母线电压幅值数据

Fig.10 Field data of voltage amplitude of a bus

表4 不同方法对现场数据状态辨识结果对比

Table 4 Comparison of the identification results of different methods on the field data states

图11 现场实测数据恢复结果

Fig.11 Recovery results of field data

参考文献 25

[1]	汤涌, 郭强, 周勤勇, 等. 特高压同步电网安全性论证[J]. 电网技术, 2016,40(1):97-104. doi: 10.13335/j.1000-3673.pst.2016.01.014 URL
	TANG Yong, GUO Qiang, ZHOU Qinyong, et al. Security evaluation for UHV synchronized power grid[J]. Power System Technology, 2016,40(1):97-104. doi: 10.13335/j.1000-3673.pst.2016.01.014 URL
[2]	申洪明, 肖智宏, 刘颖, 等. 电力电子化背景下交流电网保护配置方案[J]. 电力建设, 2020,41(7):75-81.
	SHEN Hongming, XIAO Zhihong, LIU Ying, et al. Research on AC protection scheme under the background of power electronics[J]. Electric Power Construction, 2020,41(7):75-81.
[3]	周孝信, 陈树勇, 鲁宗相. 电网和电网技术发展的回顾与展望: 试论三代电网[J]. 中国电机工程学报, 2013,33(22):1-11.
	ZHOU Xiaoxin, CHEN Shuyong, LU Zongxiang. Review and prospect for power system development and related technologies: A concept of three-generation power systems[J]. Proceedings of the CSEE, 2013,33(22):1-11.
[4]	吴双, 胡伟, 张林, 等. 基于AI技术的电网关键稳定特征智能选择方法[J]. 中国电机工程学报, 2019,39(1):14-21.
	WU Shuang, HU Wei, ZHANG Lin, et al. An intelligent key feature selection method of power grid based on artificial intelligence technology[J]. Proceedings of the CSEE, 2019,39(1):14-21.
[5]	徐兵, 张岩, 田方媛, 等. 利用有限相量测量单元的故障识别与广域后备保护策略[J]. 电力建设, 2016,37(2):25-33. doi: 10.3969/j.issn.1000-7229.2016.02.004 URL
	XU Bing, ZHANG Yan, TIAN Fangyuan, et al. Fault identification and wide-area backup protection strategy based on limited phasor measurement units[J]. Electric Power Construction, 2016,37(2):25-33. doi: 10.3969/j.issn.1000-7229.2016.02.004 URL
[6]	高峰, 刘广一, Chris Saunders, 等. 智能电网大数据的分析与应用(英文)[J]. 电力建设, 2015,36(10):11-19. doi: 10.3969/j.issn.1000-7229.2015.10.002 URL
	GAO Feng, LIU Guangyi, SAUNDERS C, et al. Applications of smart grid big data analytics[J]. Electric Power Construction, 2015,36(10):11-19. doi: 10.3969/j.issn.1000-7229.2015.10.002 URL
[7]	张素香, 赵丙镇, 王风雨, 等. 海量数据下的电力负荷短期预测[J]. 中国电机工程学报, 2015,35(1):37-42. doi: 10.13334/j.0258-8013.pcsee.2015.01.005 URL
	ZHANG Suxiang, ZHAO Bingzhen, WANG Fengyu, et al. Short-term power load forecasting based on big data[J]. Proceedings of the CSEE, 2015,35(1):37-42.
[8]	崔金栋, 王胜文, 辛业春. 区块联盟链视角下智能电网数据管理技术框架研究[J]. 中国电机工程学报, 2020,40(3):836-848.
	CUI Jindong, WANG Shengwen, XIN Yechun. Research on technical framework of smart grid data management from consortium blockchain perspective[J]. Proceedings of the CSEE, 2020,40(3):836-848.
[9]	HUANG C, LI F X, ZHOU D, et al. Data quality issues for synchrophasor applications Part I: A review[J]. Journal of Modern Power Systems and Clean Energy, 2016,4(3):342-352. doi: 10.1007/s40565-016-0217-4 URL
[10]	KIM D I, CHUN T Y, YOON S H, et al. Wavelet-based event detection method using PMU data[J]. IEEE Transactions on Smart Grid, 2017,8(3):1154-1162. doi: 10.1109/TSG.2015.2478421 URL
[11]	FAN L L, WEHBE Y. Extended Kalman filtering based real-time dynamic state and parameter estimation using PMU data[J]. Electric Power Systems Research, 2013,103:168-177. doi: 10.1016/j.epsr.2013.05.016 URL
[12]	ZHAO J B, ZHANG G X, SCALA M L, et al. Enhanced robustness of state estimator to bad data processing through multi-innovation analysis[J]. IEEE Transactions on Industrial Informatics, 2017,13(4):1610-1619. doi: 10.1109/TII.2016.2626782 URL
[13]	HUANG C, LI F X, ZHAN L W, et al. Data quality issues for synchrophasor applications Part II: Problem formulation and potential solutions[J]. Journal of Modern Power Systems and Clean Energy, 2016,4(3):353-361. doi: 10.1007/s40565-016-0213-8 URL
[14]	刘灏, 杨智伟, 毕天姝, 等. 基于优先级分配策略的PMU数据恢复方法[J]. 电网技术, 2018,42(9):2814-2820.
	LIU Hao, YANG Zhiwei, BI Tianshu, et al. PMU data recovery method based on priority allocation strategy[J]. Power System Technology, 2018,42(9):2814-2820.
[15]	杨挺, 李扬, 何周泽, 等. 基于矩阵填充的泛在电力物联网电能质量数据修复算法[J]. 电力系统自动化, 2020,44(2):13-21.
	YANG Ting, LI Yang, HE Zhouze, et al. Matrix completion theory based recovery algorithm for power quality data in ubiquitous power Internet of Things[J]. Automation of Electric Power Systems, 2020,44(2):13-21.
[16]	乔文俞, 李野, 刘浩宇, 等. 基于曲线相似与低秩矩阵的缺失电量数据补全方法[J]. 电力建设, 2020,41(1):32-38.
	QIAO Wenyu, LI Ye, LIU Haoyu, et al. Missing load data completion based on curve similarity and low-rank matrix[J]. Electric Power Construction, 2020,41(1):32-38.
[17]	OSIPOV D, CHOW J H. PMU missing data recovery using tensor decomposition[J]. IEEE Transactions on Power Systems, 2020,35(6):4554-4563. doi: 10.1109/TPWRS.59 URL
[18]	MIRANDA V, KRSTULOVIC J, KEKO H, et al. Reconstructing missing data in state estimation with autoencoders[J]. IEEE Transactions on Power Systems, 2012,27(2):604-611. doi: 10.1109/TPWRS.2011.2174810 URL
[19]	王守相, 陈海文, 潘志新, 等. 采用改进生成式对抗网络的电力系统量测缺失数据重建方法[J]. 中国电机工程学报, 2019,39(1):56-64.
	WANG Shouxiang, CHEN Haiwen, PAN Zhixin, et al. A reconstruction method for missing data in power system measurement using an improved generative adversarial network[J]. Proceedings of the CSEE, 2019,39(1):56-64.
[20]	陈继林, 陈勇, 田芳, 等. 基于LSTM算法的电网仿真样本生成方法[J]. 中国电机工程学报, 2019,39(14):4129-4135.
	CHEN Jilin, CHEN Yong, TIAN Fang, et al. The method of sample generation for power grid simulation based on LSTM[J]. Proceedings of the CSEE, 2019,39(14):4129-4135.
[21]	杨甲甲, 刘国龙, 赵俊华, 等. 采用长短期记忆深度学习模型的工业负荷短期预测方法[J]. 电力建设, 2018,39(10):20-27.
	YANG Jiajia, LIU Guolong, ZHAO Junhua, et al. A long short term memory based deep learning method for industrial load forecasting[J]. Electric Power Construction, 2018,39(10):20-27.
[22]	TAN M, YUAN S P, LI S H, et al. Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning[J]. IEEE Transactions on Power Systems, 2020,35(4):2937-2948. doi: 10.1109/TPWRS.59 URL
[23]	ERGEN T, KOZAT S S. Online training of LSTM networks in distributed systems for variable length data sequences[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018,29(10):5159-5165. doi: 10.1109/TNNLS.2017.2770179 URL
[24]	KASAC J, DEUR J, NOVAKOVIC B, et al. A conjugate gradient-based BPTT-like optimal control algorithm with vehicle dynamics control application[J]. IEEE Transactions on Control Systems Technology, 2011,19(6):1587-1595. doi: 10.1109/TCST.2010.2084088 URL
[25]	YANG Z W, LIU H, BI T S, et al. An adaptive PMU missing data recovery method[J]. International Journal of Electrical Power & Energy Systems, 2020,116:105577. doi: 10.1016/j.ijepes.2019.105577 URL