Application of multi-factor dynamic interaction graph in vulnerability assessment and online monitoring of transmission lines

Abstract

The use of interaction graph to evaluate and obtain vulnerable transmission lines is an effective means of preventing cascading failure. However, the continuous changes of the key parameters that affect the cascading failure during the operation of the power system will affect the actual vulnerability of the lines. Especially with the increase in the penetration rate of renewable energy, the impact of frequent uncertainty and randomness brought by it can no longer be ignored. Therefore, the assessment results obtained from the static interaction graph cannot meet the accuracy requirements in the engineering, and also limit the scope of its application. How to realize the dynamization of the static interaction graph is the key to solving the above problems, which is also the difficulty that the current research has been facing. This paper proposes a construction method of multi-factor dynamic interaction graph based on generalized fault chains. With data-driven as the core, it realizes true dynamic and more accurate vulnerability assessment with the help of parallel distributed computing, uninterrupted data stream and sliding window. On this basis, according to the characteristics of generalized fault chains and the dynamic interaction graph, we propose and realize online monitoring for the classification of vulnerability lines for the first time. This function provides strong support for the implementation of measures with real-time characteristics, and can also more effectively reduce the risk and harm of cascading failure. Finally, this paper combines the IEEE 118 node system and deliberate attacks to prove the significance and effectiveness of the research.