A Response-based Method For Transient Stability Analysis Using Network Structure Preserving Energy Function

With widespread application of renewable energy generation and power electronic equipment in power system,the uncertainty of power system is rapidly increasing.,and the existing event-based security and stability control mode is more and more difficult to meet the needs of power grid safe operation.On the other hand,with the rapid development of instant communication technology and wide-area measurement technology,it is possible to judge the stability simply by using the real-time response data of the system.In the field of power system transient stability analysis,a method is urgently needed to directly judge transient stability by using system fault response data without relying on off-line simulation.This thesis focuses on the idea of analyzing transient stability directly using real-time response information of the system.The main work of this thesis is summarized as follows:(1)Based on the injection energy phenomenon that during the fault the values of transient energy function continue to rise,it has been clear about the injection energy phenomena in physics that the nature of injection energy is caused by the difference in the selection of the potential energy's reference frame,and it does not exist the actual physical energy growth.In this thesis,the expressions of the structure-preserving energy function corresponding to the mathematical models of different equipment are deduced,and the mathematical terms corresponding to the injected energy phenomenon under different expressions are clarified,thus the scenarios applicable to different expressions are clarified,and the simplest expressions of the structure-preserving energy function to calculate the transient energy are deduced.Finally,a simple transient energy calculation process is proposed by discussing the expressions of the structure-preserving energy function written according to the source,network and load of the system and the expressions of the structure-preserving energy function written according to bus number of the system.(2)In the process of artificial intelligence algorithm evaluation of critical energy,the advantages of using the long-short term memory network are analyzed through the characteristics of the input data in the evaluation link,and the basic unit structure and working principle of the long-short term memory network are introduced.Based on the whole process of fault scenarios,the characteristics of change in values of the transient energy function is discussed,and how to combine the link of transient energy calculation and critical energy evaluation is clarified.Finally a complete flow chart of the method for system transient stability is deduced,and the influence of each link's output precision to the accuracy of the proposed method is derived,thus a precondition is proposed for better performance of the method.The simulation on New England 10 machine system confirms the rapidity and accuracy of the proposed method.