基于高斯混合模型的电动汽车用电行为随机模拟与充电同时系数测算方法

孙充勃, 原凯, 宋毅, 高枝, 丁羽頔, 杜孟珂

1.国网经济技术研究院有限公司,北京;2.国网北京朝阳供电公司,北京

摘要：为解决电动汽车充电站扩容规划问题,提升充电桩利用率,提出一种同时系数计算方法。采用高斯混合模型和蒙特卡洛算法的排队模型模拟用户充电行为,计算最优车桩比和同时系数。研究多场景要素对同时系数的影响,构建居民区与商业区不同场景的充电设施设计框架,通过分析私人乘用车充电功率特性曲线等测算不同场景下的充电同时系数。算例分析结果表明所提方法具有有效性和可行性,可为电动汽车充电站规划建设提供有效参考。

关键词: 电动汽车, 充电设施, 高斯混合模型, 蒙特卡洛, 充电同时系数

Stochastic Simulation of Electric Vehicle Charging Be‍havior and Charging-Simultaneity-Factor Calculation Method Based on Gaussian Mixture Model

SUN Chongbo, YUAN Kai, SONG Yi, GAO Zhi, DING Yudi, DU Mengke

1. State Grid Economic and Technological Research Institute Co., Ltd., Beijing, China;2. State Grid Beijing Chaoyang Power Supply Company, Beijing, China

Abstract: To address the expansion planning of electric vehicle(EV) charging stations and improve the utilization of charging piles, a method for calculating the simultaneity factor is proposed. A Gaussian mixture model(GMM) and a Monte Carlo algorithm-based queuing model is used to simulate user charging behavior, thereby calculating the optimal vehicle-to-pile ratio and simultaneity factor. The influence of multi-scenario factors on the simultaneity factor is investigated, and a charging facility design framework for different scenarios in residential and commercial areas is established. By analyzing charging power characteristic curves of private vehicles, the charging simultaneity factor for diverse scenarios is measured. Case study results demonstrate the effectiveness and feasibility of the proposed method, providing valuable insights for the planning and construction of EV charging stations.

Key words: electric vehicle(EV), charging infrastructure, Gaussian mixture model(GMM), Monte Carlo, charging simultaneity factor

摘自《湖南电力》2025年第六期