基于局部均值分解与神经网络的短期负荷预测  

Power System Short-term Load Forecasting Based on Local Mean Decomposition and Artificial Neural Network

作　　者：兰华[1] 常家宁[1] 周凌[1] 王冰[1] 张镭 

机构地区：[1]东北电力大学电气工程学院,吉林吉林132012 [2]长春市建设工程交易中心信息部,长春130000

出　　处：《电测与仪表》2012年第5期48-51,84,共5页Electrical Measurement & Instrumentation

摘　　要：短期负荷预测是电力系统调度和运行的基础,为了提高电力系统短期负荷预测的精度,提出了基于局部均值分解和人工神经网络的电力系统短期负荷预测方法。该方法首先对负荷序列进行局部均值分解,针对分解后具有不同特点的各PF分量设定具体的神经网络参数进行预测,将各分量的预测结果进行重构得到最终的预测结果。仿真实验表明,LMD-BP神经网络的预测方法与传统的EMD-BP神经网络方法相比具有更高的预测精度,同时也验证了该方法的实用性和有效性。Short-term load forecasting is the basis of the power system dispatching and operation. In order to improve the short-term power load precision, a novel approach for short-term load forecasting is presented based on local mean decomposition （LMD） and artificial neural network （ANN）. First of all, based on LMD the load series is decomposed into different lots of series, then according to the features of decomposed components different dynamic neural network model, finally using the BP network to reconstruct the forecasted signals of the components and obtain the ultimate forecasting result. The simulation results show that the LMD-BP neural network method has higher precision of prediction than the EMD-BP neural network method, also verify the feasibility and efficiency of this method.

关 键 词：短期负荷预测 局部均值分解 人工神经网络 

分 类 号：TM715[电气工程—电力系统及自动化]