Study Of Cascading Failures And Its Control In Power Grid Based On Structure And Power Flow Model

In today's fast-developing information age,the role of power networks is becoming more and more important.With the increasing demand for power,the scale of power networks is becoming larger and larger,so how to ensure the stable operation of a power grid and avoid large-scale cascade failures when the power grid is attacked has become an important issue.Based on the complex network theory,this dissertation conducts an in-depth study on the mechanism of cascade failure on the power grid and the method of reducing the scale of cascade failure through the structure and power flow model.The research results of this dissertation have certain guiding significance for reducing cascade failures of power grids.The specific research content is as follows:1.Research on cascading failure of power grid under capacity load model considering neighbor degree information.Considering the degree information of a node and its neighbor nodes,we propose a new node capacity load model,which is applied to the study of cascading failures in the western United States power grid and other grids.We attack the power grid by the highest load node(HL attack)and the lowest load node(LL attack).Study shows that the power grid with the improving load model has better robustness and resistance to destruction comparing with the traditional capacity load model based on the node degree.Through studying the ratio of the sum of the neighboring degree value of the attacked node to its own degree value,we explain the reason why our model can reduce the scale of cascading failures in the network.2.Research on cascade failures control of power grid based on load shedding power flow model.Under the power flow model,a new control method to reduce cascade failures of a power grid is proposed.The load-cut method adopted in this dissertation is to reduce the load of a load node by reducing the rated injection current of the load node,and finally achieve the purpose of reducing the scale of the cascade failures in the power grid.When a certain line of the power grid fails,all load nodes in the network are traversed to find the optimal load shedding node.Research shows that if the load shedding method can effectively reduce the initial cascade failures scale of the power grid,the final cascade failures size of the power grid can be reduced to a largeextent.In order to improve the search speed and efficiency of load-shedding nodes,a method for finding general solutions of load-shedding nodes in the range of 3 distances from the vertex of the initial fault line is proposed.Through simulation experiments in IEEE39,IEEE118 and IEEE300 networks,it can be found that the load shedding can effectively reduce the scale of cascading failures in the network.The method of finding a general solution can quickly locate the load node that need to be cut,and effectively reduce the scale of cascading failures in a network.3.Key lines identification of a power grid based on a greedy algorithm.Under the power flow model of the power grid,a series of important lines in the power grid are found by introducing a greedy algorithm.The greedy algorithm attack method means that each attack can cause the largest scale of cascading failures in the current surviving network.By comparing with the line importance attack method and the line current attack method,it is found that the greedy algorithm attack mode can crash the power grid at the fastest speed by removing the least number of lines.This proves the effectiveness of the greedy algorithm attack mode,and a series of attacked lines based on the greedy algorithm are important lines in the power grid.