A Fast Searching Method for Cascading Failure Pattern Based on PrefixSpan Algorithm

Studying the propagation mechanism of cascading failure and identifying the vulnerable lines in the power grid is of great significance for the prevention control of cascading failure. The existing works pay more attention to vulnerable lines and high-risk path, which are hard to explain the overall propagation characteristic of cascading failure. This paper applies sequential pattern mining technology to cascading failure analysis. The concept of cascading failure pattern (CFP) is defined firstly, and a cascading failure pattern mining algorithm (CFPMA) based on PrefixSpan is proposed. The relevance of lines is then calculated based on the result of mining algorithm. Test result on the IEEE 39-bus system shows that the proposed method can find out the CFP which can clearly show the overall propagation mechanism of cascading failure. Furthermore, test results under different power flow snapshots indicate that the CFP doesn't change with system status.