| 引用本文: | 刘金晨,陈文钢,陈佳佳.基于期望成本最小化聚类与动态识别的光储充电站最优运行[J].电力自动化设备,2025,45(10):151-159. |
| LIU Jinchen,CHEN Wengang,CHEN Jiajia.Optimal operation of photovoltaic-energy storage charging station based on minimized expected cost-based clustering and dynamic identification[J].Electric Power Automation Equipment,2025,45(10):151-159. |
|
| 摘要: |
| 针对电动汽车(EV)充电需求不确定性所导致的光储充电站运行策略难以制定的问题，提出一种基于期望成本最小化聚类与动态识别的光储充电站最优运行策略。建立考虑价格型需求响应的EV充电负荷预测模型，兼顾EV充电负荷的随机性和自主选择性。为了避免频繁调整光储充电站的运行策略，采用期望成本最小化的聚类方法对EV充电负荷进行聚类，并根据生成的EV充电负荷曲线优化对应的光储充电站运行策略。综合考虑取值范围、斜率相似性、功率差值相似性3个指标，采用多特征提取方法对任意日EV充电负荷曲线所属类进行动态识别，确定光储充电站的日内运行策略。算例分析结果表明，所提策略可有效提升光储充电站的收益和运行效率。 |
| 关键词: 光储充电站 电动汽车 随机聚类 期望成本最小 动态识别 需求响应 运行策略 |
| DOI：10.16081/j.epae.202506021 |
| 分类号: |
| 基金项目:国家自然科学基金资助项目(52377110) |
|
| Optimal operation of photovoltaic-energy storage charging station based on minimized expected cost-based clustering and dynamic identification |
|
LIU Jinchen, CHEN Wengang, CHEN Jiajia
|
|
School of Electrical and Electronic Engineering, Shandong University of Technology, Zibo 255000, China
|
| Abstract: |
| Aiming at the problem that it is difficult to formulate the operation strategy of photovoltaic-energy storage charging station caused by the uncertainty of electric vehicle(EV) charging demand, an optimal operation strategy of photovoltaic-energy storage charging station based on minimized expected cost-based cluste-ring and dynamic identification is proposed. The EV charging load forecasting model considering price-based demand response is constructed, which takes both the randomness and autonomous selectivity of EV charging load into consideration. In order to avoid frequently adjusting the operation strategy of photovoltaic-energy storage charging station, a minimized expected cost-based clustering method is adopted to categorize the EV load, and the corresponding operation strategies of photovoltaic-energy storage charging station are optimized based on the generated EV load curves. Comprehensively considering three indicators, namely the value range, slope similarity and power difference similarity, the category that any daily EV load curve belongs to is dynamically identified by using the multi-feature extraction method, and the intra-day operation strategy of photovoltaic-energy storage charging station is determined. The results of the case study analysis show that the proposed strategy can effectively improve the revenue and operational efficiency of the photovoltaic-energy storage charging station. |
| Key words: photovoltaic-energy storage charging station electric vehicles stochastic clustering minimized expected cost dynamic identification demand response operation strategy |
