摘要
依据数据挖掘理论对数据进行收集、整合,运用改进型BP神经网络模型处理数据,建立电力负荷模型进行短期预测.通过不同精度下的实验分析,结果表明,改进型神经网络负荷预测模型在高精度下预测结果优于低精度下预测结果,最大误差同比降低80%,适用实际负荷预测.
Based on data mining theory to collect and integrate data,using improved BP neu ral network model to process the grid data,and a electrical load model was esta blished to forcast the short-term electrical load.Through different precision experiments,the high-precision forcast result is better than the low-precisio n forcast result,the maximum error is reduced by 80%,it is proved that the mod el suits to the practical.
出处
《华北水利水电学院学报》
2011年第3期43-45,共3页
North China Institute of Water Conservancy and Hydroelectric Power
关键词
数据挖掘
改进BP算法
人工神经网络
电力负荷预测
L-M法
data mining
modified BP algorithm
Artificial Neural Networks
electrical load forecast
Levernberg-Marquardt algorithm.
