Stability Detection of DC Microgrid System with Passive Damping Based on XGboost Algorithm

In the DC microgrid, constant power loads (CPL) experience negative damping characteristics. This phenomenon often leads to resonance during the cascading process, thereby affecting the stability of the system. To solve the issue, this paper proposes a stability detection strategy for DC microgrid systems with passive damping, using the extreme gradient boosting (XGboost) method. Firstly, passive damping is incorporated into the DC microgrid filter, and the impedance model of the DC microgrid system is constructed to identify the parameters that influence system stability. Subsequently, these parameters are utilized as variables to establish a simulation scenario, acquiring sample data for training the XGBoost. Finally, the effectiveness of the XGBoost method in DC microgrid system stability detection is evaluated using the confusion matrix. Simulation results verify the effectiveness of the proposed method.