王强,陈浩,刘炼

• 1:三峡大学电气与新能源学院

摘要(Abstract)：

为提升电压稳定裕度预测的精准度和增强预测模型的可解释性，将自然梯度提升算法与沙普利值加性解释理论应用于电压稳定裕度预测中。以离线样本为基础，采用自然梯度提升算法探索运行状态变量与电压稳定裕度间的非线性映射关系，构建自然梯度提升驱动的电压稳定裕度预测模型。然后，引入沙普利值加性解释理论对自然梯度提升模型进行解释，构建基于沙普利值加性解释理论的电压稳定裕度预测影响因素分析架构，并通过全局分析和个体分析两个角度，给出各特征量对于电压稳定裕度预测的具体影响过程，挖掘出导致系统电压稳定裕度降低的关键因素。在新英格兰39节点系统上的算例分析结果表明，与其他算法相比，自然梯度提升不仅具有最佳的预测精度，还拥有良好的鲁棒性与泛化能力，并且基于沙普利值加性解释理论的影响因素分析架构为电压稳定裕度预测提供了依据和支撑。

关键词(KeyWords)： 电压稳定裕度;机器学习;自然梯度提升;沙普利值加性解释;可解释性

Abstract:

Keywords:

基金项目(Foundation): 国网江西省电力有限公司科技项目（5218F0180049）

作者(Author): 王强,陈浩,刘炼

DOI: 10.19635/j.cnki.csu-epsa.000942

参考文献(References)：

• [1]Fan Youping,Liu Songkai,Qin Libin,et al.A novel online estimation scheme for static voltage stability margin based on relationships exploration in a large data set[J].IEEETrans on Power Systems,2015,30(3):1380-1393.
• [2]林伟芳，易俊，郭强，等(Lin Weifang,Yi Jun,Guo Qiang,et al).阿根廷“6.16”大停电事故分析及对中国电网的启示(Analysis on blackout in Argentine power grid on June 16,2019 and its enlightenment to power grid in China)[J].中国电机工程学报(Proceedings of the CSEE),2020,40(9):2835-2841.
• [3]Zhang Xiaoping,Ju Ping,Handschin E.Continuation three-phase power flow:a tool for voltage stability analysis of unbalanced three-phase power systems[J].IEEE Trans on Power Systems,2005,20(3):1320-1329.
• [4]Lee D H A.Voltage stability assessment using equivalent nodal analysis[J].IEEE Trans on Power Systems,2015,31(1):454-463.
• [5]刘昇，徐政，华文，等(Liu Sheng,Xu Zheng,Hua Wen,et al).用于在线预测静态电压稳定性的SIPSS-Lasso-BP网络(A SIPSS-Lasso-BP network for online forecasting static voltage stability)[J].中国电机工程学报(Proceedings of the CSEE),2014,34(34):6032-6041.
• [6]唐滢淇，董树锋，朱承治，等(Tang Yingqi,Dong Shufeng,Zhu Chengzhi,et al).基于Tri-Training-LASSO-BP网络的静态电压稳定裕度在线预测方法(Online prediction method of static voltage stability margin based on Tri-Training-LASSO-BP network)[J].中国电机工程学报(Proceedings of the CSEE),2020,40(12):3824-3834.
• [7]Wang Bo,Fang Biwu,Wang Yajun,et al.Power system transient stability assessment based on big data and the core vector machine[J].IEEE Trans on Smart Grid,2016,7(5):2561-2570.
• [8]Huang Jiyu,Guan Lin,Su Yinsheng,et al.A topology adaptive high-speed transient stability assessment scheme based on multi-graph attention network with residual structure[J].International Journal of Electrical Power&Energy Systems,2021,130:106948.
• [9]Ren Chao,Xu Yan.Transfer learning-based power system online dynamic security assessment:using one model to assess many unlearned faults[J].IEEE Trans on Power Systems,2020,35(1):821-824.
• [10]丁长新，张沛，孟祥飞，等(Ding Changxin,Zhang Pei,Meng Xiangfei,et al).基于分类回归树算法的在线静态电压稳定裕度评估(Online evaluation on static voltage stability margin based on classification and regression tree algorithm)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2020,32(1):93-100.
• [11]和怡，朱小军，李登峰，等(He Yi,Zhu Xiaojun,Li Dengfeng,et al).基于模态分析和Relief算法的在线静态电压稳定特征选取方法(Feature selection method for online static voltage stability based on modal analysis and Relief algorithm)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2017,29(7):87-92.
• [12]肖繁，王涛，饶渝泽，等(Xiao Fan,Wang Tao,Rao Yuze,et al).基于梯度提升决策树静态电压稳定裕度评估(Static voltage stability margin evolution based on gradient boosting regression tree)[J].电测与仪表(Electrical Measurement&Instrumentation),2020,57(20):39-45.
• [13]王慧芳，张晨宇(Wang Huifang,Zhang Chenyu).采用极限梯度提升算法的电力系统电压稳定裕度预测(Prediction of voltage stability margin in power system based on extreme gradient boosting algorithm)[J].浙江大学学报(工学版)(Journal of Zhejiang University(Engineering Science)),2020,54(3):606-613.
• [14]梁修锐，刘道伟，杨红英，等(Liang Xiurui,Liu Daowei,Yang Hongying,et al).数据驱动的电力系统静态电压稳定态势评估(Data-driven situation assessment of power system static voltage stability)[J].电力建设(Electric Power Construction),2020,41(1):126-132.
• [15]卢锦玲，郭鲁豫(Lu Jinling,Guo Luyu).基于改进深度残差收缩网络的电力系统暂态稳定评估(Power system transient stability assessment based on improved deep residual shrinkage network)[J].电工技术学报(Transactions of China Electrotechnical Society),2021,36(11):2233-2244.
• [16]陈明华，刘群英，张家枢，等(Chen Minghua,Liu Qunying,Zhang Jiashu,et al).基于XGBoost的电力系统暂态稳定预测方法(XGBoost-based algorithm for post-fault transient stability status prediction)[J].电网技术(Power System Technology),2020,44(3):1026-1034.
• [17]Duan T,Avati A,Ding D Y,et al.NGBoost:natural gradient boosting for probabilistic prediction[EB/OL].https：//arxiv.org/abs/1910.03225,2020.
• [18]Lundberg S M,Lee Su-In.A unified approach to interpreting model predictions[C]//31st Annual Conference on Neural Information Processing Systems.Long Beach,USA,2017:4766-4755.
• [19]Su Heng-Yi,Liu Tzu-Yi.Enhanced-online-random-forest model for static voltage stability assessment using wide area measurements[J].IEEE Trans on Power Systems,2018,33(6):6696-6704.
• [20]Liu Songkai,Shi Ruoyuan,Zhang Tao,et al.An integrated scheme for static voltage stability assessment based on correlation detection and random bits forest[J].International Journal of Electrical Power&Energy Systems,2021,130:106898.
• [21]Ke Guolin,Meng Qi,Finley T,et al.Light GBM:a highly efficient gradient boosting decision tree[C]//31st Annual Conference on Neural Information Processing Systems.Long Beach,USA,2017:3147-3155.
• [22]He Kaiming,Zhang Xiangyu,Ren Shaoqing,et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas,USA,2016:770-778.