基于广义回归神经网络的月度负荷预测

Monthly Load Forecasting Based on General Regression Neural Network

研究了月度负荷的特性，指出了其季节波动性和趋势增长性双重特性；介绍了广义回归神经网络的基本理论，提出以横向历史数据和纵向历史数据作为输入神经元，建立了月度负荷预测模型，并将其应用于我国某地区月度负荷预测，结果表明：该模型的预测精度明显高于一般的BP网络。

After studying the characteristic of monthly history load, the paper point out monthly load is possessed of the property of increase trend and seasonal fluetuation simultaneously,so its behavior appears as the characteristic of complex non - linearity. The paper presents the basics of a general regression neural network （GRNN）, use the horizontal and vertical history data as input of GRNN, building the monthly load forecasting model , apply GRNN to monthly load forecast. The forecasting results shows GRNN is more accurate than back propagation （BP）.