Dominant Assessment of Transient Instability in Power System Based on Integrated Convolutional Neural Networks
For the coupling between voltage instability and power angle instability during transient instability,traditional power system transient stability assessment methods find it difficult to quantify the two instability modes,thus making it difficult to reasonably select corresponding transient stability control strategies.This is a problem that needs to be addressed.This article proposes a method for dominant instability assessment based on integrated convolutional neural networks.This method utilizes the transient energy function method to con-struct sensitivity indicators,with the dominant unstable state of the system as the output of the neural net-work.Three different convolutional neural network models are constructed and the model by selecting input features for training is optimized.The simulation and analysis of the IEEE-39 node example by using this mod-el verifies that the method has high robustness in the absence of generator nodes.The application of this mod-el is beneficial for improving the stability and safety of the power system.
transient stability of power systemintegrated convolutional neural networktransient dominant in-stability
万方数据
