Travel Behavior Simulation of Electric Vehicle Cluster Based on Gaussian Mixture Model

In recent years, the rapid development of electric vehicles (EVs) has led to a significant increase in load, gradually exerting a greater impact on the safe and stable operation of the power grid. Therefore, it is of paramount importance to simulate EV travel behaviors and acquire EV loads. However, many existing studies suffer from issues such as oversimplified models and low accuracy. To address this, this paper proposes an EV behavior simulation method based on Gaussian Mixture Models (GMM) and the Bayesian principle, which demonstrate broad applicability and high robustness, yielding fitting results with high precision. Firstly, this paper describes the Gaussian Mixture Model and the Bayesian principle employed. Subsequently, it sequentially simulates trip chain variables such as the frequency of EV trips, trip durations, trip destinations, and parking times, while implementing specific treatments for continuous variables. Finally, through case studies, this paper validates the effectiveness of the algorithm.