| 引用本文: | 顾 睿,于艾清,潘含芝,等.基于改进贝叶斯最大熵的乡村旅游电动汽车多时间尺度充电负荷预测[J].电力系统保护与控制,2025,53(12):117-127.[点击复制] |
| GU Rui1,YU Aiqing,PAN Hanzhi,et al.Multi-timescale charging load forecasting for rural tourism electric vehicles based on improved Bayesian maximum entropy[J].Power System Protection and Control,2025,53(12):117-127[点击复制] |
|
| 摘要: |
| 当前乡镇电动汽车(electric vehicle, EV)充电负荷样本集的获取受限于充电网络覆盖率低，给乡村旅游EV充电负荷预测带来极大挑战。且目前研究大多局限于短期预测范畴，鲜有涉及对多时间尺度的深入探讨。基于此，提出一种基于改进贝叶斯最大熵(Bayesian maximum entropy, BME)的乡村旅游EV多时间尺度充电负荷预测模型。首先，考虑EV的出行特性受温度与交通因素的影响建立EV单位能耗模型, 在此基础上建立基于改进BME的乡村旅游EV短期负荷预测模型。其次，结合最优灰色模型与旅游客流量预测模型预测未来乡村旅游EV保有量，从而推演出乡村旅游EV中长期负荷预测结果。最后，基于江苏省某乡村旅游景区温度与行车数据进行仿真分析， 验证所提方法的有效性并预测乡村旅游EV充电负荷的未来发展趋势。 |
| 关键词: 电动汽车 多时间尺度负荷预测 贝叶斯最大熵 季节性特征 |
| DOI：10.19783/j.cnki.pspc.241061 |
| 投稿时间：2024-08-08修订日期：2024-11-22 |
| 基金项目:上海市自然科学基金项目资助(23ZR1425000)；上海市科技创新行动计划项目资助(22010501400) |
|
| Multi-timescale charging load forecasting for rural tourism electric vehicles based on improved Bayesian maximum entropy |
| GU Rui11,YU Aiqing1,PAN Hanzhi2,YANG Feixiang1,WANG Yufei1,XUE Hua1 |
| (1. College of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China;
2. East China Electric Power Design Institute Co., Ltd., Shanghai 200063, China) |
| Abstract: |
| The limited availability of electric vehicle (EV) charging load data in rural areas, due to low charging network coverage, poses a significant challenge for forecasting EV charging loads in rural tourism settings. Furthermore, existing studies mainly focus on short-term forecasting, with limited exploration of multi-timescale predictions. To address these issues, an enhanced multi-timescale charging load forecasting model for rural tourism EVs is proposed based on an improved Bayesian maximum entropy (BME) framework. First, an EV unit energy consumption model is established considering temperature and traffic impacts on EV travel behavior. Based on this, a short-term load forecasting model using the improved BME in developed. Next, by integrating an optimal grey model with tourist flow prediction, future stock of rural tourism EVs is forecasted, thereby deriving medium- and long-term charging load forecasts. Finally, simulation analysis using temperature and traffic data from a rural tourism area in Jiangsu Province is conducted to validate the effectiveness of the proposed method and predict the future development trend of rural tourism EV charging loads. |
| Key words: electric vehicles multi-timescale load forecasting Bayesian maximum entropy seasonal characteristics |
