Resilience-oriented distributed generation planning of distribution network under multiple extreme weather conditions

Planning resilient and elastic distribution networks has become an effective strategy for withstanding extreme weather events. However, conventional research often overlooks the impact of multiple extreme weather conditions and only considers planning for one specific type of extreme weather. This paper proposes a resilience-oriented distributed generation planning approach for distribution networks, which takes into account multiple extreme weather conditions. Firstly, a line fault probability model is established to capture the impact of typhoons, rainstorms, ice and snow weather. Secondly, a fault scenario generation and simplification method based on modified monte carlo simulation and k-means clustering is proposed to ensure the representativeness and computational efficiency of the selected scenarios. Additionally, a two-stage stochastic mixed integer programming model is introduced to enhance system resilience through distributed generation configuration and network topology reconstruction. The first stage focuses on determining the number, location, and size of distributed generation with economic objectives, while the second stage addresses the recovery method after an uncertain extreme event based on the distributed generation configuration obtained in the first stage. The proposed model is applied to modified IEEE 33-bus system and IEEE 123-bus system. Compared with the conventional method, the DG configuration results are consistent and the solution time is reduced by 70–80 %, achieving a balance between accuracy and efficiency. Furthermore, it effectively reduces load shedding by nearly 90 % by optimizing DG utilisation.