Transient Stability Feature Selection Method Based on Deep Learning Technology
- 36 Downloads
As a key link in the transient stability assessment of power system, feature selection is the basis to ensure the transient stability assessment results. In view of the defects in the methods proposed in the existing literatures at home and abroad, this paper proposes a deep learning model adapted to the feature extraction of power grid simulation data, which is based on the deep topology convolutional network to extract features. Simulation results show that the proposed model has very high reliability for network stability, and the obtained characteristic quantity can be effectively connected with the data analysis algorithm and achieve good results.
KeywordsDeep learning Graph convolution network (GCN) Transient stability assessment (TSA) Fast stability determination
- 1.Zhang C, Li Y, Yu Z, et al. A weighted random forest approach to improve predictive performance for power system transient stability assessment. In: 2016 power and energy engineering conference (APPEEC). Xi’an: IEEE; 2016. p. 1259–63.Google Scholar
- 2.Shengyong Y, et al. Study on power systems transient stability assessment based on machine learning method. Chengdu: Southwest Jiaotong University; 2010.Google Scholar
- 5.Qian M, Yihan Y, Wenying L, et al. Power system transient stability assessment with combined SVM method mixing multiple input features. Proc CSEE. 2005;25(6):17–23.Google Scholar
- 6.Ye S, Wang X, Liu Z, et al. Transient stability assessment based on random forest algorithm. J Southwest Jiaotong Univ. 2008;43(5):573–7.Google Scholar