基于EMD和相关向量机的短期负荷预测

为提高电力负荷预测的准确性,提出一种基于经验模态分解EMD(empirical mode decomposition)与相关向量机RVM(relevant vector machine)的短期负荷预测方法。该方法利用EMD将目标负荷序列分解为若干个不同频率的固有模态分量I MF(intrinsic mode function),通过分析各个分量的特征规律,构造不同的RVM模型对各分量分别进行预测,再将各分量预测值通过RVM组合得到最终预测值。仿真结果表明,通过EMD分解,预测效果有显著改善,而RVM模型较之BP神经网络模型与SVM模型具有更高的预测精度。

Short-Term Load Forecasting Based on EMD and RVM

In order to improve the accuracy of power system load forecasting,a load forecasting model based on empirical mode decomposition(EMD)and relevant vector machine(RVM)is proposed.Using the EMD,this method decomposed the target load sequence into a number of different frequency components of the intrinsic mode function(IMF).By analyzing the characteristics of various components of the law,different models to forecast each component separately were consturcted using RVM,and then these forecasting results of eac...