Power System Transient Stability Assessment Based on Voltage Phasor and Convolution Neural Network

With the increasing complexity of power systems, the existing transient stability assessment methods have many shortcomings in using power grid interconnection dynamic information. In response to these deficiencies, a new method for power system transient stability assessment is presented based on voltage phasor and CNN(Convolution Neural Network). Firstly, using the deep learning techniques which is widely applied in the field of image processing, a dynamic display of power system transient process in the voltage phasor complex plane is constructed. Secondly, based on CNN and the image of voltage phasor complex plane, the power system transient stability fast evaluation model is proposed. Additionally, to solve the problem of sample imbalance, a multi-window sliding recognition method is proposed to supplement the sample. Finally, the data model is applied on the IEEE 10 Machine 39 bus system to verify the validity and accuracy of the method. This method solves the problem of feature selection and sample imbalance in the process of transient fault analysis based on data, and provides some reference for parameters setting of deep learning model.