Management of Cascading Outage Risk Based on Risk Gradient and Markovian Tree Search

Since cascading outages are major threats to power system operations, it is of great significance to reduce the risk of potential cascading outages. In this paper, a method for reduction of cascading outage risk based on Markovian tree (MT) search is proposed. Based on the MT expansion of the cascading outage risk, the risk gradient is computed with a forward-backward tree search scheme. The computation of risk gradient is incorporated into the procedure of risk assessment based on MT search, which is efficient with good convergence. Then the an optimization model for risk reduction (RR) is formulated by using risk gradient, which minimizes the cost of control while effectively reduces the cascading outage risk. Moreover, to overcome the limitation of linearization, an iterative risk reduction (IRR) algorithm is further developed. Test results on a 4-bus test system and the RTS-96 3-area test system verify the accuracy of risk gradient computation and effectiveness of the RR. And the performance of the IRR is demonstrated on RTS-96 3-area system and a 410-bus US-Canada northeast system model. The results show that the subsequent cascade risk and the total risk are reduced by 93.6% and 54.5%, respectively.