MP-EVData：中国多原型电动汽车充电负载曲线的人工智能增强数据集 负载曲线在中国

电动汽车（EV）的大规模融合给 电网运行带来了重大挑战操作，需要高保真、 多样化集来进行更深入的研究。为了满足这一需求，我们推出了 MP-EVData，这是一个综合数据集，由 组成 中国主要大都市的充电站级充电负载曲线组成，涵盖整个 2024 日历年。 的核心核心价值MP-EVData充电负荷数据10个站时间段内代表五个不同的原型：出租车示范站、公交站、住宅充电站、换电站和重型卡车站。这种独特的结构消除了的 等外部变量地理、气候和政策受控分析。此外，该数据集还增加了使用先进的生成式人工智能模型生成的并行、高保真合成数据集，以支持数据密集型研究。技术验证揭示了原型中高度不同的每日、每周和每年时间模式，并展示了在使用时间定价下明确的价格响应充电行为。MP-EVData 为推进埃斯方面的电动汽车负载预测、智能充电算法和城市基础设施规划The large-scale integration of electric vehicles (EVs) poses significant challenges to power grid operations, necessitating high-fidelity and diverse datasets for in-depth research. To meet this demand, we present MP-EVData, a comprehensive dataset comprising station-level charging load curves from major Chinese megacities, covering the full 2024 calendar year. The core value of MP-EVData lies in its charging load data spanning 10 station time periods, which represents five distinct prototype scenarios: taxi service stations, bus charging stations, residential charging stations, battery swap stations, and heavy-duty truck charging stations. This unique structure eliminates external variables such as geography, climate, and policy from controlled analytical studies. Additionally, this dataset is supplemented with parallel, high-fidelity synthetic datasets generated using advanced generative artificial intelligence models to support data-intensive research. Technical validations have revealed highly distinct daily, weekly, and annual temporal patterns across these prototypes, and demonstrate clear price-responsive charging behavior under time-of-use pricing schemes. MP-EVData facilitates advancements in EV load forecasting, intelligent charging algorithm development, and urban infrastructure planning.

搜集汇总

背景与挑战

背景概述

MP-EVData是一个综合性的中国电动汽车充电负载曲线数据集，包含2024年五种不同原型充电站的真实数据和AI生成的合成数据，适用于电动汽车负载预测和智能充电算法等研究。