With the increasing coupling between transient angle and voltage stability issues in power systems, the demand for the integrated assessment of high-dimensional redundant features has become prominent. This study proposes a two-stage feature selection method: first, key features were screened using class separability scores and Pearson correlation coefficients to eliminate classification inefficiency and redundancy. Subsequently, a binary competitive search algorithm was employed to optimize the feature subset, and the assessment model was constructed combined with a multi-task convolutional gated recurrent network. Case studies demonstrate that this method significantly improves transient stability assessment effectiveness while reducing feature dimensionality.
WANG H Y,CHEN Q F,ZHANG B H.Transient stability assessment combined model framework based on cost-sensitive method[J].IET Generation,Transmission & Distribution,2020,14(12): 2256-2262.
ZHU L P,HILL D J,LU C.Intelligent short-term voltage stability assessment via spatial attention rectified RNN learning[J].IEEE Transactions on Industrial Informatics,2021,17(10): 7005-7016.
ZHANG Y C,XU Y,DONG Z Y,et al.A hierarchical self-adaptive data-analytics method for real-time power system short-term voltage stability assessment[J].IEEE Transactions on Industrial Informatics,2019,15(1): 74-84.