Hydrogen refueling siting: A case study from China on the influence of commercial entities

Abstract

This study introduces machine-learning techniques to analyze the impact of surrounding commercial entities on hydrogen refueling station (HRS) site selection. A large-scale multi-entity dataset is established through field research and data fusion, and the random forest (RF) algorithm is used to quantify the importance of influencing factors, thereby overcoming the subjectivity bias in existing studies. The cross-validation results suggest that the RF model has high stability and generalizability for HRS site selection. In addition, the RF model excels in classification tasks and maintains consistent performance across different datasets. This study provides valuable insights into HRS site selection by incorporating commercial entity data and leveraging machine-learning techniques.