XGBoost-based Algorithm for Post-fault Transient Stability Status Prediction

被引：0

作者：

Chen M. ^{[1
]}

Liu Q. ^{[1
]}

Zhang J. ^{[1
]}

Chen S. ^{[1
]}

Zhang C. ^{[1
]}

机构：

[1] School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu, 611731, Sichuan

来源：

Dianwang Jishu/Power System Technology | 2020年 / 44卷 / 03期

基金：

中国博士后科学基金; 中国国家自然科学基金;

关键词：

Feature importance; Model interpretation; PMU; Transient stability prediction; XGBoost;

D O I：

10.13335/j.1000-3673.pst.2018.1649

中图分类号：

学科分类号：

摘要：

For transient stability prediction of power systems based on artificial intelligence method, the prediction model obtained based on a large number of samples in off-line training is often a 'black box', making the model with poor interpretation. In this paper, a transient stability prediction model based on extreme gradient boosting (XGBoost) was constructed using the operating characteristics of the generators after power system fault. The model had a great advantage in the trade-off between prediction accuracy and calculation speed. Experiment results on New England 39-bus system demonstrated that the proposed model had better accuracy and less time consumption compared to other methods. At the same time, the model provided feature importance scores and decision graph, so as to discover the relationship between features and transient stability. Finally, an algorithm explaining the prediction results for a specific fault was proposed, further improving interpretability. © 2020, Power System Technology Press. All right reserved.

引用

下载

页码：1026 / 1033

页数：7

共 19 条

[1] Liu D., Song D., Wang H., Et al., Voltage stability online evaluation system based on WAMS and EMS, Power System Technology, 38, 7, pp. 1934-1938, (2014)
[2] Wu W., Tang Y., Sun H., Et al., A Survey on research of power system transient stability based on wide-area measurement information, Power System Technology, 36, 9, pp. 81-87, (2012)
[3] Zhou D.Q., Annakkage U.D., Rajapakse A.D., Online monitoring of voltage stability margin using an artificial neural network, IEEE Transactions on Power Systems, 25, 3, pp. 1566-1574, (2010)
[4] Tang B., Deng C., Liu L., Application of compound neural network in power system transient stability assessment, Power System Technology, 28, 15, pp. 62-66, (2004)
[5] Gomez F.R., Rajapakse A.D., Annakkage U.D., Et al., Support vector machine-based algorithm for post-fault transient stability status prediction using synchronized measurements, IEEE Transactions on Power Systems, 26, 3, pp. 1474-1483, (2011)
[6] Ye S., Wang X., Liu Z., Et al., Power system transient stability assessment based on support vector machine incremental learning method, Automation of Electric Power Systems, 35, 11, pp. 15-19, (2011)
[7] Zhou Y., Wu J., Yu Z., Et al., Support vector machine ensemble classifier and its confidence evaluation for transient stability prediction of power systems, Power System Technology, 41, 4, pp. 1188-1196, (2017)
[8] Wang Y., Wang B., Tang F., Et al., Power system on-line transient stability assessment based on response trajectory and core vector machine, Proceedings of the CSEE, 34, 19, pp. 3178-3186, (2014)
[9] Dai Y., Chen L., Zhang W., Et al., Power system transient stability assessment based on multi-support vector machines, Proceedings of the CSEE, 36, 5, pp. 1173-1180, (2016)
[10] Huang Y., Yu Z., Shi D., Et al., Strategy of huge electric power system stability quick judgment based on massive historical online data, Proceedings of the CSEE, 36, 3, pp. 596-603, (2016)

← 1 2 →