论文部分内容阅读
开展电动汽车充电负荷预测研究是引导电动汽车进行有序充放电的基础。依据当前广东某电动公交充电站的相关实测数据,对电动公交充电站负荷特征进行分析,提出一种基于小波神经网络(WNN)的电动公交站短期负荷预测方法。利用该方法对随机选取的两组测试日进行预测实例分析,并与单一BP网络模型的预测效果进行比较。统计结果表明,基于WNN的预测方法具有较高的预测精度,满足一定的应用要求,适用于电动公交充电站短期负荷预测。
Carrying out the study of electric vehicle charging load forecasting is the basis to guide the electric vehicle to orderly charging and discharging. According to the current measured data of an electric bus charging station in Guangdong, the load characteristics of the electric bus charging station are analyzed and a short-term load forecasting method based on wavelet neural network (WNN) is proposed. This method is used to make a case study of two randomly selected test days and compared with that of a single BP network model. The statistical results show that the WNN-based prediction method has high prediction accuracy and meets certain application requirements and is suitable for short-term load forecasting of electric bus charging stations.