Demand-Side Virtual Power Plant External Characterization Parameters and Correlation Model Taking into Account Yunnan Resource Characteristics

Aiming at the problem of accurately obtaining and optimizing the external characteristic parameters of the demand-side Virtual Power Plant (VPP) in Yunnan Province, this paper designs a classification and optimization method based on decision tree and random forest. The effectiveness of the proposed method is verified through simulation experiments, in which the decision tree model performs feature selection and node partitioning through information entropy and Mantaras paradigm distance, while the random forest model utilizes integrated learning to improve the accuracy of prediction and the generalization ability of the model. Experimental results show that the method can accurately classify nine typical resources with 99.5% accuracy. In addition, through feature importance analysis, the minimum and average values were identified as the most representative feature parameters. This study provides a powerful decision support for the optimization and stable operation of the power system in Yunnan Province, and can accurately describe the external characteristic parameters of VPP.