Analysis of Online Quick Judgment of Transient Stability Based on Siamese Network

Dynamic security assessment (DSA) is widely used in dispatching operation systems, and calculation speed is one of its most important performance indices. In this paper, a deep learning model called Siamese neural network is proposed aiming to predict the transient stability indicators of power system, for example critical clearing time (CCT). The method is much faster than the simulation and suitable for online analysis. Firstly, a simulation sample database is constructed based on historical online data; then a Siamese model is trained, which uses static state quantities as its inputs like active power of generators. While a new online power flow needs to be evaluated, the high level features of Siamese model are obtained and a k-NN is implemented to find the most familiar samples in the database using the chosen features; the final result will be determined comprehensively by the familiar samples. The validity of proposed method is verified by the simulation using online data of State Grid Corp of China (SGCC) and different key faults. It is proved that the method meets the requirements for speed and accuracy of online analysis system, especially for small sample set.