A Sensitivity Analysis Method for Distribution Network Investment Based on Adaptive Parallel Sampling

With the rapid integration of novel elements such as distributed power sources, electric vehicles, and new energy storage systems, along with novel business forms like microgrids, into modern power systems, it has become increasingly important to evaluate their impact on distribution network investment. This study proposes a sensitivity analysis method that combines the XGBoost agent model with the EFAST sensitivity analysis to quantify such impacts. First, initial data sampling is performed using the EFAST method. Then, the relationship between the development of novel elements and business forms and the associated investment costs is captured through the XGBoost agent model. Finally, the sampling space is iteratively optimized using a parallel sampling strategy, enhancing the accuracy of sensitivity analysis. A case study based on real-world data validates the proposed method, demonstrating its effectiveness in identifying and quantifying the influence of emerging technologies and business models on distribution network investment.