Please wait a minute...
浙江大学学报(工学版)  2020, Vol. 54 Issue (3): 606-613    DOI: 10.3785/j.issn.1008-973X.2020.03.022
电气工程     
采用极限梯度提升算法的电力系统电压稳定裕度预测
王慧芳*(),张晨宇
浙江大学 电气工程学院,浙江 杭州 310027
Prediction of voltage stability margin in power system based on extreme gradient boosting algorithm
Hui-fang WANG*(),Chen-yu ZHANG
Department of Electrical Engineering, Zhejiang University, Hangzhou 310027, China
 全文: PDF(773 KB)   HTML
摘要:

将极限梯度提升树(XGBoost)算法应用于电力系统电压稳定评估问题. 根据电压稳定问题特点,提出能够反映电力系统运行状态的特征集;把电压稳定裕度绝对值作为映射目标,并介绍生成样本集的方法. 在介绍XGBoost算法基本原理的基础上,研究该算法的技术细节. 在IEEE-39节点系统上进行验证,结果表明,XGBoost算法在R方值和平均绝对百分误差2项回归指标上均优于其他几类机器学习算法,且模型的计算速度最快,可以满足在线应用要求. 同时,XGBoost算法具有良好的数值错误和数值缺失容错性,并可以针对预测偏差较大的样本进行数据补充,实现模型的更新,使得模型表现趋于稳定.

关键词: 电力系统电压稳定性机器学习人工智能极限梯度提升树(XGBoost)算法    
Abstract:

The extreme gradient boosting (XGBoost) algorithm was applied in power system voltage stability assessment problem. According to the characteristics of the voltage stability problem, a feature set which could reflect the state of a power system was defined. Taking the absolute value of voltage stability margin as the mapping target, the method to generate the sample set was studied. Based on the introduction of the basic principle of the XGBoost algorithm, the technical details of the algorithm were discussed. The algorithm was evaluated in the IEEE-39 power system. Results show that the XGBoost algorithm has better performance than other machine learning models according to two evaluation metrics: R squared value and mean absolute percentage error value, and has the fastest computation speed, which can meet the demand of online application. Meanwhile, the XGBoost algorithm is proved to be robust when the data errors and data missing happen. And data supplement can be taken for the samples with large prediction deviation to update the model, thus making the performance of the model more stable.

Key words: power system    voltage stability    machine learning    artificial intelligence    extreme gradient boosting (XGBoost) algorithm
收稿日期: 2019-03-03 出版日期: 2020-03-05
CLC:  TM 744  
通讯作者: 王慧芳     E-mail: huifangwang@zju.edu.cn
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  
王慧芳
张晨宇

引用本文:

王慧芳,张晨宇. 采用极限梯度提升算法的电力系统电压稳定裕度预测[J]. 浙江大学学报(工学版), 2020, 54(3): 606-613.

Hui-fang WANG,Chen-yu ZHANG. Prediction of voltage stability margin in power system based on extreme gradient boosting algorithm. Journal of ZheJiang University (Engineering Science), 2020, 54(3): 606-613.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2020.03.022        http://www.zjujournals.com/eng/CN/Y2020/V54/I3/606

图 1  电压稳定裕度
特征符号 特征含义
V iθ i 节点i 的电压幅值、相角
PGiQGi 发电机节点i 的有功、无功出力
PLiQLi 负荷节点i 的有功、无功负荷需求
PBijQBij 线路ij传输的有功、无功功率
表 1  电网稳态特征
图 2  IEEE-39节点系统示意图
模型 R2 MAPE
KNNRegressor 0.946 4.869
SVR 0.961 4.152
RF 0.977 2.960
GBRT 0.987 2.140
XGBoost 0.992 1.621
表 2  不同机器学习模型的预测表现对比
图 3  不同机器学习算法的预测效果
模型 Tt /s Tp /s
KNNRegressor 0.022 0 0.480 6
SVR 1.753 2 0.405 5
RF 9.060 5 0.006 0
GBRT 7.285 5 0.003 2
XGBoost 4.609 0 0.000 1
表 3  不同模型训练、预测的消耗时间对比
误差范围 R2 MAPE 误差范围 R2 MAPE
5% 0.972 3.096 1% 0.988 2.002
2% 0.980 2.478 0.5% 0.991 1.789
表 4  样本存在数值错误时的XGBoost模型预测效果
缺失比例 R2 MAPE
1% 0.987 1.905
2% 0.976 2.356
表 5  样本存在数值缺失时XGBoost模型效果
图 4  测试集误差最大的50个样本的真实值和预测偏差
XGBoost模型 样本集 $\emptyset $ 测试集
MAE MRE% R2 MAPE
更新前 219 4.460 0.992 1.621
更新后 2.49 0.017 0.996 1.313
表 6  XGBoost模型更新前、后效果对比
1 王锡凡. 现代电力系统分析[M]. 北京: 科学出版社, 2003.
2 RODRIGUES A B, PRADA R B, SILVA M D G D Voltage stability probabilistic assessment in composite systems: modeling unsolvability and controllability loss[J]. IEEE Transactions on Power Systems, 2010, 25 (3): 1575- 1588
doi: 10.1109/TPWRS.2009.2039234
3 AJJARAPU V, CHRISTY C The continuation power flow: a tool for steady state voltage stability analysis[J]. IEEE Transactions on Power Systems, 1992, 7 (1): 416- 423
doi: 10.1109/59.141737
4 杨滢, 叶琳, 倪秋龙 基于PSS/E的浙江电网静态电压稳定性分析[J]. 浙江电力, 2009, 28 (6): 9- 11
YANG Ying, YE Lin, NI Qiu-long Analysis of static voltage stability of Zhejiang electric grid based on PSS/E[J]. Zhejiang Electric Power, 2009, 28 (6): 9- 11
doi: 10.3969/j.issn.1007-1881.2009.06.003
5 吴杰康, 邓松, 梁志武, 等 基于模糊神经网络决策树的电压稳定性评估[J]. 电网技术, 2007, 32 (14): 25- 30
WU Jie-kang, DENG Song, LIANG Zhi-wu, et al Evaluation of power system voltage stability based on fuzzy neural network decision tree[J]. Power System Technology, 2007, 32 (14): 25- 30
6 王皓, 孙宏斌, 张伯明 基于混合互信息的特征选择方法及其在静态电压稳定评估中的应用[J]. 中国电机工程学报, 2006, 26 (7): 77- 81
WANG Hao, SUN Hong-bin, ZHANG Bo-ming Hybrid mutual information based feature selection method as appied to static voltage stability assessment in power systems[J]. Proceedings of the CSEE, 2006, 26 (7): 77- 81
doi: 10.3321/j.issn:0258-8013.2006.07.015
7 FAN Y, LIU S, QIN L, et al A novel online estimation scheme for static voltage stability margin based on relationships exploration in a large data set[J]. IEEE Transactions on Power Systems, 2015, 30 (3): 1380- 1393
doi: 10.1109/TPWRS.2014.2349531
8 DEVARAJ D, ROSELYN J P On-line voltage stability assessment using radial basis function network model with reduced input features[J]. International Journal of Electrical Power and Energy Systems, 2011, 33 (9): 1550- 1555
9 崔峰, 齐占庆, 姜萌 基于模糊神经网络的电力系统电压稳定评估[J]. 电力系统保护与控制, 2009, 37 (11): 40- 44
CUI Feng, QI Zhan-qing, JIANG Meng Fuzzy neural network based voltage stability evaluation of power systems[J]. Power System Protection and Control, 2009, 37 (11): 40- 44
doi: 10.3969/j.issn.1674-3415.2009.11.010
10 刘昇, 徐政, 华文 用于在线预测静态电压稳定性的SIPSS-Lasso-BP网络[J]. 中国电机工程学报, 2014, (34): 6032- 6041
LIU Sheng, XU Zheng, HUA Wen A SIPSS-Lasso-BP network for online forecasting static voltage stability[J]. Proceedings of the CSEE, 2014, (34): 6032- 6041
11 ZHOU D Q, ANNAKKAGE U D, RAJAPAKSE A D Online monitoring of voltage stability margin using an artificial neural network[J]. IEEE Transactions on Power Systems, 2010, 25 (3): 1566- 1574
doi: 10.1109/TPWRS.2009.2038059
12 FAN Y, LI X, ZHANG P Real-time static voltage stability assessment in large-scale power systems based on maximum-relevance minimum-redundancy ensemble approach[J]. IEEE Access, 2017, 5: 27281- 27291
doi: 10.1109/ACCESS.2017.2758819
13 MALBASA V, ZHENG C, CHEN P C, et al Voltage stability prediction using active machine learning[J]. IEEE Transactions on Smart Grid, 2017, 8 (6): 3117- 3124
doi: 10.1109/TSG.2017.2693394
14 CHEN T, GUESTRIN C. XGBoost: a scalable tree boosting system [C] // Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining: SIGKDD. San Francisco: ACM, 2016: 785-794.
15 WANG H, ZHANG C, LIN D et al An artificial intelligence-based method for evaluating power grid node importance using network embedding and support vector regression[J]. Frontiers of Information Technology and Electronic Engineering, 2019, 20 (6): 816- 828
doi: 10.1631/FITEE.1800146
16 ZIMMERMAN, R D, MURILLO-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
17 PEDREGOSA F, GRAMFORT A, MICHEL V, et al Scikit-learn: machine learning in python[J]. Journal of Machine Learning Research, 2016, 12 (10): 2825- 2830
[1] 战友,李强,马啸天,王郴平,邱延峻. 基于宏微观纹理特征融合的路面摩擦性能预测[J]. 浙江大学学报(工学版), 2021, 55(4): 684-694.
[2] 于勇,薛静远,戴晟,鲍强伟,赵罡. 机加零件质量预测与工艺参数优化方法[J]. 浙江大学学报(工学版), 2021, 55(3): 441-447.
[3] 陈巧红,陈翊,李文书,贾宇波. 多尺度SE-Xception服装图像分类[J]. 浙江大学学报(工学版), 2020, 54(9): 1727-1735.
[4] 康庄,杨杰,郭濠奇. 基于机器视觉的垃圾自动分类系统设计[J]. 浙江大学学报(工学版), 2020, 54(7): 1272-1280.
[5] 谢乐,衡熙丹,刘洋,蒋启龙,刘东. 基于线性判别分析和分步机器学习的变压器故障诊断[J]. 浙江大学学报(工学版), 2020, 54(11): 2266-2272.
[6] 万志远,陶嘉恒,梁家坤,才振功,苌程,乔林,周巧妮. Stack Overflow上机器学习相关问题的大规模实证研究[J]. 浙江大学学报(工学版), 2019, 53(5): 819-828.
[7] 柯懂湘,潘丽敏,罗森林,张寒青. 基于随机森林算法的Android恶意行为识别与分类方法[J]. 浙江大学学报(工学版), 2019, 53(10): 2013-2023.
[8] 忽丽莎, 王素贞, 陈益强, 高晨龙, 胡春雨, 蒋鑫龙, 陈振宇, 高兴宇. 基于可穿戴设备的跌倒检测算法综述[J]. 浙江大学学报(工学版), 2018, 52(9): 1717-1728.
[9] 王洪凯, 陈中华, 周纵苇, 李迎辞, 陆佩欧, 王文志, 刘宛予, 于丽娟. 机器学习算法诊断PET/CT纵膈淋巴结性能评估[J]. 浙江大学学报(工学版), 2018, 52(4): 788-797.
[10] 吴鹏洲,于慧敏,曾雄. 基于正则化风险最小化的目标计数[J]. 浙江大学学报(工学版), 2014, 48(7): 1226-1233.
[11] 王冠楠,孙黎滢,甘德强,王彬彬,辛焕海. 电力系统稳定器设计的广义相位补偿法[J]. 浙江大学学报(工学版), 2014, 48(7): 1295-1303.
[12] 吕文韬,沈忱,江道灼,桂帆,范宇,吴兆麟. 具有电容限压功能的限流式统一潮流控制器[J]. 浙江大学学报(工学版), 2014, 48(5): 877-881.
[13] 杨波, 钟彦儒, 曾光. 阶梯波链式静止同步补偿器电容电压平衡控制[J]. J4, 2014, 48(4): 600-609.
[14] 阎博,江道灼,甘德强,藏玉清. 基于反馈线性化H∞方法的UPFC非线性鲁棒控制器[J]. J4, 2012, 46(11): 1975-1980.
[15] 王康, 符杨, 辛焕海, 王冠楠. 基于新型Back-stepping方法的电力系统
励磁控制器设计
[J]. J4, 2011, 45(4): 747-753.