Transient security state identification of smart grid based on multi feature fusion

Abstract

In order to improve the power supply stability of the smart grid and accurately identify the transient safety status of the power grid, a smart grid transient safety status identification method based on multi feature fusion is proposed. Firstly, extract the transient zero sequence active energy features of the smart grid, and use the S transform to extract the transient energy features and comprehensive phase angle features. Secondly, based on the extracted multiple features, a deep belief network (DBN) is used to fuse multiple features. Finally, based on the results of multi feature fusion, the SVM algorithm is used to classify and identify the transient safety status of the power grid. The experimental results show that the transient safety state identification accuracy of this method is high, stable at 98%; and the misjudgement rate of this method has been reduced, with a maximum of no more than 3%.