基于鹈鹕优化算法优化广义回归神经网络的电动汽车充电负荷短期预测  

Electric Vehicle Charging Load Short-term Prediction Based on Generalized Regression Neural Network Optimized by Pelican Optimization Algorithm

作　　者：陈晓华 吴杰康[2] 张勋祥 龙泳丞 王志平 CHEN Xiaohua;WU Jiekang;ZHANG Xunxiang;LONG Yongcheng;WANG Zhiping(Zhanjiang Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Zhanjiang 524005,China;School of Automation,Guangdong University of Technology,Guangzhou 510006,China;School of Electrical Engineering&Intelligentization,Dongguan University of Technology,Dongguan 523808,China)

机构地区：[1]广东电网有限责任公司湛江供电局,广东湛江524005 [2]广东工业大学自动化学院,广东广州510006 [3]东莞理工学院电子工程与智能化学院,广东东莞523808

出　　处：《山东电力技术》2024年第7期1-9,共9页Shandong Electric Power

基　　金：国家自然科学基金项目(50767001)。

摘　　要：针对目前电动汽车充电负荷预测精度不足的问题,提出了一种结合互补集合经验模态分解和鹈鹕优化算法优化广义回归神经网络的组合预测方法。首先,利用互补集合经验模态分解将电动汽车充电负荷时间序列分解成多个固有模态函数分量和一个残差分量。其次,对于分解后的固有模态分量容易出现冗杂信息,利用样本熵对分解后数值相近的固有模态分量进行相加重构,降低冗杂程度。最后,考虑广义回归神经网络的预测效果与平滑因子的数值有很大关系,利用鹈鹕优化算法优化广义回归神经网络的平滑因子,进而对电动汽车充电负荷进行短期预测。仿真表明,所提出的预测方法可以有效地提高电动汽车充电负荷的预测精度,具有较高的实用性。To solve the problem of insufficient prediction accuracy of electric vehicle charging load,a combined prediction method combining complementary ensemble empirical mode decomposition and pelican optimization algorithm to optimize generalized regression neural network was proposed.Firstly,the time series of electric vehicle charging load was decomposed into multiple intrinsic mode function components and a residual component by using complementary ensemble empirical mode decomposition.Then,considering that the decomposed intrinsic mode components are prone to redundant information,sample entropy was used to add and reconstruct the decomposed intrinsic mode components with similar values to reduce the degree of redundancy.Finally,considering that the prediction effect of the generalized regression neural network is closely related to the value of the smoothing factor,the pelican optimization algorithm was used to optimize the smoothing factor of the generalized regression neural network,and then the short-term prediction of the electric vehicle charging load was carried out.The simulation results show that the proposed prediction method can effectively improve the prediction accuracy of electric vehicle charging load with high practicability.

关 键 词：广义回归神经网络 鹈鹕优化算法 电动汽车充电负荷 短期预测 互补集合经验模态分解 

分 类 号：TM714[电气工程—电力系统及自动化]