Open AccessTransient Voltage Stability Assessment Method based on gcForest
Author(s) -
Zhen Sun,
Mingpo Li,
Jie Zhang,
Hu Binjiang,
Qi Guo,
Yicheng Zhu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1914/1/012025
Subject(s) - transient (computer programming) , transient voltage suppressor , computer science , stability (learning theory) , voltage , hyperparameter , electric power system , artificial neural network , control theory (sociology) , power (physics) , transformation (genetics) , transient response , artificial intelligence , engineering , machine learning , electrical engineering , chemistry , control (management) , operating system , biochemistry , physics , quantum mechanics , gene
Aiming at the fact that the traditional method cannot quickly and accurately judge the power system transient voltage stability, and the deep neural network method needs many sample data sets and has a large number of hyperparameters, a power system transient voltage stability assessment method based on deep forest (gcForest) is proposed. gcForest effectively extracts features through ensemble learning, multi-layer feature transformation, and dynamically adjusting the number of model layers. Finally, this method is applied to transient voltage stability assessment of an electrolytic aluminium load-intensive area in Guangxi Power Grid. The results show that the method has high accuracy, the feasibility and effectiveness of this method are verified. This method can assist dispatchers to judge the transient voltage stability of the power grid.