A Combined Clustering Scenario Reduction Method Base on Potential Decision Results for Generation Expansion Planning

Driven by the carbon peaking and carbon neutrality goals, the capacity of renewable energy (RE) will begin to increase by leaps and bounds. The uncertainty of RE will challenge the operation verification of generation expansion planning (GEP). In this paper, a combined clustering method based on potential decision results is proposed for scenario reduction (SR) to reduce the scale of operational constraints. By solving the GEP model based on one single scenario, the multidimensional input domain is transformed into the potential decision result domain. The SOM and k-medoids clustering methods are combined to reduce the scenario while retaining the advantages of both methods. Then the XJTU-ROTS system is used as an example to verify the effectiveness and advantages of the method. The results show that compared with traditional clustering methods, the proposed method has the best SR effect under the optimal number of clusters.