An Improved Robust Optimization Method for Resilient Distribution Networks with the Strategies of Line Hardening and Distributed Generator Placement

Jian Wang,Zepeng Li,Tong Qian,Wenhu Tang

DOI: https://doi.org/10.1109/acpee53904.2022.9783704

2022-01-01

Abstract:In recent years, a variety of natural disasters have caused large area blackouts in the power distribution network. Therefore, it is fundamental to research how to enhance the resilience of the distribution network. This paper proposes an improved robust optimization method for resilient distribution networks with the strategies of line hardening and distributed generator placement. The problem is formulated as a tri-level mixed-integer model, the first-level defender develops planning strategies based on the worst-case scenario decided by the second-level attacker, and the system will be redispatched in the third level to minimize the load shedding. By setting a confidence level, worst-case scenarios with low probability will be eliminated from the model, which improves the conservativeness compared to the conventional robust optimization model. A customized column-and-constraint generation algorithm is proposed to solve this nonlinear problem by decomposing the proposed model into two linear models and a non-linear constraint. The effectiveness and economy of the proposed model is verified based on a modified IEEE 33-bus distribution test system. This result can provide the power company with a more economical planning strategy compared to the conventional robust optimization model under a given budget.