Power System Transient Voltage Stability Assessment Based on Kernel Principal Component Analysis and DBN

Aiming at the transient voltage stability assessment of power system, this paper proposes a method based on kernel principal component analysis (KPCA) and deep belief network (DBN) for power system transient voltage stability assessment. Firstly, a set of 45-dimensional eigenvectors that can reflect the transient voltage stability of the power system is constructed. The feature vector set is reduced based on KPCA, the feature vector dimension and the filtering redundancy feature are reduced, then the dimensionality-reduced feature vector is input into the DBN network. The training process consists of pre-training and fine-tuning is performed to optimize the DBN grid structure parameters. The simulation results of the 10-machine 39-node in New England show that the method can reduce the dimension of input data, remove redundant features, reduce the error rate, and test time of transient stability assessment. It can accurately and quickly judge the steady state voltage state of the power system.