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OA

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.