基于CPS和VMD的实时电力需求饱和自适应预测

Adaptive prediction of real⁃time power demand saturation based on CPS and VMD

电力数据具有非平稳性和随机性的特点,导致电力需求预测结果误差较大,为此,提出基于CPS和VMD的实时电力需求饱和自适应预测方法。基于CPS与VMP建立自适应变分模态分解模型,利用实时电力需求序列和各模态分解余量之间的信息散度差来判断相关性程度,实现模态分解,通过自适应模态分解得到电力需求序列的分解余量,依靠其建立自适应预测模型,并对模型进行训练,完成对实时电力需求饱和自适应预测。实验结果表明,设计方法预测结果的平均绝对误差低于5.0,平均绝对百分比误差低于2.45%。

Power data is characterized by nonstationarity and randomness,which leads to large errors in power demand forecasting results.Therefore,a real-time adaptive forecasting method for power demand saturation based on CPS and VMD is proposed.An adaptive variational modal decomposition model is established based on CPS and VMP,and the degree of correlation is judged by the difference of information divergence between the real-time power demand sequence and each modal decomposition margin to realize modal decomposition.The decomposition margin of the power demand sequence is obtained through the adaptive modal decomposition,and an adaptive prediction model is established based on it,and the model is trained to complete the adaptive prediction of the real-time power demand saturation.The experimental results show that the average absolute error of the prediction results of the design method is less than 5.0,and the average absolute value percentage error is less than 2.45%.