摘要
针对电动汽车充电负荷随机性强、易受气象条件影响以及采用传统ARIMA模型预测准确度不高等问题,提出了一种考虑气象因素的改进ARIMA充电负荷预测模型。采用窗口滚动的方式更新数据并计算模型参数,以保证数据源在时间尺度上的有效性;通过分析气象特征向量与负荷序列的相关度,合理分配模型参数权重因子的赋值,以克服充电负荷随天气变化呈现的突发性和随机性。最后,以北京市某充电站的实测数据为测试样本,进行了仿真分析和对比试验。结果表明,所提模型正确有效且具有较高的预测精度。
A load forecasting model considering meteorological factors based on improved ARIMA was proposed,as the EV charging load is highly random and easily affected by meteorological factors,besides the traditional ARIMA model is not accurate. In order to ensure the validity of data sources on the time scale,the model updates the data by the way of window scroll and calculates parameters,then assigns weighting factors of model parameters reasonably by analyzing the correlation between meteorological feature vector and load sequence to overcome the sudden and random changes of charging load with the weather. Finally,taking the measured data of a charging station in Beijing as the test example,a simulation analysis and contrast experiment were carried out. The results show that the proposed model is effective and has high prediction accuracy.
引文
[1]李立理,张义斌.国内外电动汽车市场的比较分析及启示[J].中国电力,2013,46(10):74-77.
[2]胡泽春,宋永华,徐智威,等.电动汽车接入电网的影响与利用[J].中国电机工程学报,2012,32(4):1-10.
[3]刘青,戚中译.基于蒙特卡洛法的电动汽车负荷预测建模[J].电力科学与工程,2014,30(10):14-19.
[4]张洪财,胡泽春,宋永华,等.考虑时空分布的电动汽车充电负荷预测方法[J].电力系统自动化,2014,38(1):13-20.
[5]黄小庆,陈颉,陈永新,等.大数据背景下的充电站负荷预测方法[J].电力系统自动化,2016,40(12):68-74.
[6]刘文霞,龙日尚,徐晓波,等.考虑数据新鲜度和交叉熵的电动汽车短期充电负荷预测模型[J].电力系统自动化,2016,40(12):45-52.
[7]常德政,任杰,赵建伟,等.基于RBF-NN的电动汽车充电站短期负荷预测研究[J].青岛大学学报(工程技术版),2014,29(4):44-48.
[8]范美强,廖维林,吴伯荣,等.电动车用MH-Ni电池温度特性研究[J].电池工业,2004,9(6):287-289.
[9]孟天星,张厚升.基于ARIMA模型的风电场短期风速预测[J].科学技术与工程,2013,13(33):27-32.
[10]DICKEY D.Time series theory and methods[J].Springer,1991,15(2):159-181.
[11]罗浩成,胡泽春,张洪财.环境温度对电动汽车充电负荷的影响分析[J].电力建设,2015,36(7):69-74.
[12]徐泰山,鲍颜红,苏寅生,等.暂态稳定断面功率极限区间和关联度指标计算[J].电力系统自动化,2016,40(20):154-160.