Fast Evaluation Of Power System Transient Stability Based On Voltage Phasor And Deep Learning

In recent years,China,the United States,India and other countries have occurred in a large area of power outages,leading to a huge economic losses and far-reaching social impact.Renewable energy,the further growth of power load,the power system of elecricity and other widely used in the power grid to increase the uncertainty and complexity of the operation of the grid safe and stable operation is facing severe challenges.With the rise of artificial intelligence and depth learning,the researchers provide a solid theoretical and practical basis for analyzing the safety and stability of grid information physical system from the perspective of large data,large sample and probability.In this paper,the CNN method is introduced into the safety and stability analysis of the power system by constructing a visualized full information grid diagram using convolution neural network(CNN).Based on the analysis of predecessors' work,the traditional method and convolution neural network method are used to analyze the security stability of the system,and the intelligent decision-making in the operation of large power grid is supported.This article mainly includes the following:(1)In order to build a visual full information grid diagram.The simulation plane of the power system based on the voltage complex plane is established.The simulation data and the existing topological connection relation are obtained with the New England 10 machine 39 nodes as an example.Based on Echart,the node information of the system is dynamically displayed on the voltage complex plane.(2)The transient stability analysis model of power system based on voltage complex plane dynamic information is established.From the perspective of weak cross section,the relationship between voltage phasor distance and oscillation center branching branch is analyzed,and the method of fast identification of weak section is given.On the basis of this,the characteristics of the phase trajectory of the busbar side busbar with weak cross section are analyzed,and the method of discriminating the transient stability based on the trajectory characteristics of the bus at both ends of the weak cross section is proposed.(3)Using the method of Deep learning to deal with data in artificial visual field,the method of depth learning is applied to the transient stability evaluation of power system in order to deal with the image pixel matrix.Based on the sample batch manufacturing program and the convolution neural network model,the depth learning model is established by the experimental method,and the multi-window sliding recognition method is proposed,which reduces the parameters of the model and improves the accuracy rate.