摘要
相比于普通电动汽车,纯电重卡具有更高的充电功率、更大的电池容量和更可观的调度潜力,同时受货物重量、物流特性及行驶路径等诸多因素影响,其充电负荷呈现更大的随机性,建立准确的纯电重卡集群负荷预测模型有利于掌握其充电规律,降低对电网的冲击。为此,提出了考虑物流特性基于高阶马尔科夫链的纯电重卡集群负荷预测方法。首先,在考虑软时间窗约束实现纯电重卡路径规划的基础上,通过分析纯电重卡行驶特性预测其开始充电时间,以获取各时刻纯电重卡充电数量;其次,对纯电重卡的荷电状态区间进行模糊双层离散化分区处理,将每个大区间进一步细分为n个小区间,提高预测精度;然后,在求取纯电重卡荷电状态多步转移概率基础上,采用高阶马尔可夫链建立集群负荷预测模型,实现更为精确的负荷预测;最后,采用某物流园区实际纯电重卡数据进行仿真验证,结果表明,所构建的负荷预测模型较准确地预测了纯电重卡集群功率,同时减小了普通马尔科夫链方法的预测误差。
Compared with ordinary electric vehicles,electric trucks have higher charging power,larger battery capacity and more considerable dispatching potential,while their charging load presents greater randomness due to many factors such as cargo weight,logistics characteristics and driving path.To this end,this paper proposes a electric trucks aggregation load prediction method based on higher-order Markov chain considering logistics characteristics.Firstly,on the basis of considering the soft time window constraint to realize the path planning of the electric trucks,the charging time of the electric trucks is predicted by analyzing their driving characteristics to obtain the charging quantity of the electric trucks at each moment.Secondly,the charge state interval of the electric trucks is partitioned with fuzzy two-level discretization,and each large interval is further subdivided into n small intervals so as to improve the prediction accuracy.And then,after obtaining the charge state multi-step transfer probability of the electric trucks,a aggregation load prediction model is established by using high-order Markov chain to achieve more accurate load prediction.Finally,the actual electric truck data of a logistics park is used for simulation verification,and the results show that the proposed load prediction model accurately predicts the aggregation power of electric trucks and reduces the prediction error of the ordinary Markov chain method.
作者
刘航
申皓
杨勇
纪陵
余洋
LIU Hang;SHEN Hao;YANG Yong;JI Ling;YU Yang(State Grid Handan Electric Power Supply Company,Handan 056000,China;Guodian Nanjing Automation Co.,Ltd.,Nanjing 210032,China;State Key Laboratory of Alternate Electrical Power System With Renewable Energy Sources(North China Electric Power University),Baoding 071003,China)
出处
《中国电力》
CSCD
北大核心
2024年第5期61-69,共9页
Electric Power
基金
国家自然科学基金资助项目(52077078)
国网河北省电力有限公司科技项目(支撑纯电重卡集群与电网友好互动的聚合调控关键技术研究及示范应用,kj2022-050)。
关键词
纯电重卡
高阶马尔可夫链
负荷预测
双层离散化
物流订单约束
electric truck
higher-order Markov chains
load forecasting
two-level discretization
logistics order constraints
