基于粒子群优化算法支持向量回归预测法的大电网电压稳定在线评估方法

doi:10.3969/j.issn.1008-0198.2022.05.004

湖南电力 ›› 2022, Vol. 42 ›› Issue (5): 22-28.doi: 10.3969/j.issn.1008-0198.2022.05.004

基于粒子群优化算法支持向量回归预测法的大电网电压稳定在线评估方法

李帅虎¹, 赵翔¹, 蒋昀宸²

1.长沙理工大学电气与信息工程学院,湖南长沙 410014;
2.国网湖南省电力有限公司长沙供电分公司,湖南长沙 410011

收稿日期:2022-07-15 修回日期:2022-07-29 出版日期:2022-10-25 发布日期:2022-11-16
作者简介:李帅虎(1981),男,博士,副教授,通信作者,主要从事电力系统电压稳定分析与控制、智能电网的自愈调控方法、大规模储能建模与优化控制方法的研究。
基金资助:
国家自然科学基金(51777179)

An On-Line Voltage Stability Assessment Method based on Particle Swarm Optimization Support Vector Regression Prediction

LI Shuaihu¹, ZHAO Xiang ¹, JIANG Yunchen²

1. School of Electrical and Information Engineering, Changsha University of Science and Technology, Changsha 410014, China;
2. State Grid Changsha Power Supply Company, Changsha 410011, China

Received:2022-07-15 Revised:2022-07-29 Online:2022-10-25 Published:2022-11-16

摘要/Abstract

摘要： 提出基于粒子群优化算法支持向量回归预测法(particle swarm optimization support vector regression, PSO-SVR)的大电网电压稳定在线评估方法,将传统基于深度神经网络(deep neural networks,DNN)模型的电压稳定评估方法改进为PSO优化过的SVR模型,对阻抗模裕度指标进行预测。该方法利用了SVR模型具有学习能力强、泛化错误率低的优点,在小样本的情况下也可以很好地学习到样本中的特征。同时克服SVR模型对于参数调节和函数选择非常敏感的问题,利用PSO算法对SVR模型的超参数进行优化选择,可以让SVR模型更好地学习到电网运行数据和阻抗模裕度值之间的非线性关系。最后,该方法在IEEE 118节点系统进行验证,并与基于DNN模型的评估方法进行比较,验证了其精度水平高于基于DNN模型的方法。

关键词: 电力系统, 静态电压稳定, 阻抗模裕度, 粒子群优化算法, 支持向量回归预测法

Abstract: In this paper, an online voltage stability evaluation method based on particle swarm optimization support vector regression(PSO-SVR) for large power grid is proposed. The traditional voltage stability evaluation method based on deep neural networks(DNN) model is improved to the SVR model optimized by PSO to predict the impedance modulus margin index.This method takes advantage of SVR model′s advantages of strong learning ability and low generalization error rate, and it can also learn the features of samples well in the case of small samples.In order to overcome SVR model is very sensitive to parameter adjustment and function selection.The proposed method uses PSO algorithm to optimize the hyper-parameters of SVR model, which can make the SVR model better learn the nonlinear relationship between power grid operation data and impedance modulus margin value.Finally, the proposed method is validated in IEEE 118 node system, and the accuracy of the proposed method is higher than that of DNN model-based evaluation method.

Key words: power system, static voltage stability, impedance modulus margin, particle swarm optimization, support vector regression

中图分类号:

TM712.2

李帅虎, 赵翔, 蒋昀宸. 基于粒子群优化算法支持向量回归预测法的大电网电压稳定在线评估方法[J]. 湖南电力, 2022, 42(5): 22-28.

LI Shuaihu, ZHAO Xiang , JIANG Yunchen. An On-Line Voltage Stability Assessment Method based on Particle Swarm Optimization Support Vector Regression Prediction[J]. Hunan Electric Power, 2022, 42(5): 22-28.

参考文献

[1]Vu K T, Liu C C, Taylor C W, et al. Voltage instability:mechanisms and control strategies [power systems][J].Proceedings of the IEEE, 1995, 83(11): 1442-1455.
[2]Van Cutsem T. Voltage instability: phenomena, countermeasures, and analysis methods [J]. Proceedings of the IEEE, 2000, 88(2): 208-227.
[3]Liu H, Chen X, Yu K, et al. The control and analysis of self-healing urban power grid [J]. IEEE Transactions on smart grid, 2012, 3(3): 1119-1129.
[4]张尧,张建设,袁世强. 求取静态电压稳定极限的改进连续潮流法[J]. 电力系统及其自动化学报,2005,17(2):21-25.
[5]马睿博. 基于深度学习的电力系统静态稳定性预测[D].北京:北京邮电大学,2018.
[6]Wang B, Fang B, Wang Y, et al. Power system transient stability assessment based on big data and the core vector machine [J]. IEEE Transactions on Smart Grid, 2016, 7(5): 2561-2570.
[7]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.
[8]Su H, Liu T. Enhanced-Online-Random-Forest Model for Static Voltage Stability Assessment Using Wide Area Measurements [J]. IEEE Transactions on Power Systems. 2018, 33(6): 6696-6704.
[9]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. 2008, 32(14): 25-30.
[10]Diao R, Sun K, Vittal V, et al. Decision Tree-Based Online Voltage Security Assessment Using PMU Measurements [J]. IEEE Transactions on Power Systems. 2009, 24(2): 832-839.
[11]Kai S, Likhate S, Vittal V, et al. An Online Dynamic Security Assessment Scheme Using Phasor Measurements and Decision Trees [J]. IEEE Transactions on Power Systems. 2007, 22(4): 1935-1943.
[12]Xie L, Chen Y, Kumar P R. Dimensionality reduction of synchrophasor data for early event detection: Linearized analysis [J]. IEEE Transactions on Power Systems, 2014, 29(6): 2784-2794.
[13]Sunitha R, Kumar R S, Mathew A T. Online static security assessment module using artificial neural networks [J]. IEEE transactions on power systems, 2013, 28(4): 4328-4335.
[14]吴倩红,韩蓓,李国杰,等. 极高光伏渗透率下基于潮流雅可比矩阵和卷积神经网络的静态电压稳定在线预测[J]. 中国电机工程学报,2021,41(12):4058-4067.
[15]晏光辉,刘颂凯,张磊,等.基于迁移学习的智能静态电压稳定评估方案[J].电测与仪表,2022,59(2):92-99.
[16]马民,秦佳,杨东升,等.人工智能在电力系统中的应用综述[J].郑州大学学报(工学版),2019,40(5):23-31.
[17]Chu S C,Du Z G,Peng Y J, et al. Fuzzy Hierarchical Surrogate Assists Probabilistic Particle Swarm Optimization for expensive high dimensional problem[J]. Knowledge-Based Systems, 2021, 220(1):106939.
[18]Li S, Hou J, Yang A, et al. DNN-Based Distributed Voltage Stability Online Monitoring Method for Large-Scale Power Grids[J]. Frontiers in Energy Research, 2021, 9: 80.

基于粒子群优化算法支持向量回归预测法的大电网电压稳定在线评估方法

An On-Line Voltage Stability Assessment Method based on Particle Swarm Optimization Support Vector Regression Prediction

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价