Research On Identification Of Critical Lines And Risk Assessment Of Cascading Failures In Power Grid

The power industry occupies an important position in the development of the national economy.At present,the scale of the power system is expanding.This improves the reliability of power supply and the economic efficiency,but also increases the risk of large-scale blackout.In recent years,there have been many large-scale blackouts in the world,all of which have brought a blow to the national economy.It is found that a large blackout event is mostly caused by a series of cascading failures.Therefore,exploring the mechanism of cascading failure,finding the critical lines in cascading failures,searching for critical lines in cascading failures'occurrence and development,and building a cascading failure risk assessment model to assess the security level of the power grid are of great significance to avoid the occurrence of blackouts.This paper first begins with the research background and significance of the subject.Several exsiting typical methods for identifying critical lines are summarized and some shortcomings of these methods are analyzed.The mechanism of cascading failure propagation is explained,and the current research methods and risk assessment methods of cascading failure are analyzed.In the aspect of critical lines identification,the PageRank algorithm in web links is introduced in this paper.Firstly,the line correlation index is constructed based on N-1 fault to characterize the directional relationship between lines.Secondly,the index of weighted power flow fluctuating entropy is proposed on the basis of the previous trend entropy index,and then the comprehensive index is established to evaluate the influence of line fault on the system.Finally,the improved PageRank algorithm is proposed,so that it can comprehensively reflect impact of the line fault on the system and the influence degree of the line affected by the fault of other lines,and more comprehensively assess critical lines in cascading failures' occurrence and development.In the aspect of power network security analysis based on the risk of cascading failure,multi-fault combinations which cause the larger risk of cascading failure need to be watched.A layer-by-layer screening method is adopted for selecting these multi-fault combinations.Firstly,the first layer fault is selected based on the critical line identification results.Secondly,the follow-up fault is selected by using the comprehensive evaluation index of the influence of line fault on system.Considering that using the Monte Carlo method to evaluate the average risk value of cascading failures caused by multiple fault combinations has time-consuming disadvantages,the Support Vector Machine(SVM)regression algorithm is proposed to quickly estimate the average risk of cascading failures to achieve real-time assessment of grid security.The above research is simulated in the IEEE 39 node system,and the result of the example is analyzed and compared to verify the effectiveness of the proposed index system and algorithm.