摘要
提出了一种基于灰色模型和神经网络组合的短期负荷预测方法。首先利用频域分解消除负荷序列的周期性,然后利用灰色模型计算负荷序列的历史拟合值和未来预测值,将其作为神经网络的输入。在历史数据中选择一天作为基准日,以该基准日的量为参照,以负荷的灰色模型拟合值相对基准日的变化量,以及温度变化量为BP神经网络的输入,实际负荷变化量为输出,训练神经网络并预测待预测日负荷的变化量,加上基准日负荷后得到预测负荷。该方法综合了灰色模型方法和神经网络方法的优点,仿真结果验证了方法的有效性。
A short-term load forecasting method is proposed based on the combination of grey model and artificial neural network. The weekly period of the load series is eliminated with the frequency decomposition method, and the grey model is used to compute the historical fitting values and future forecasting values of the load series. One day of the historical data is chosen as the reference day. The difference between the grey model fitting value and the reference day load, together with the temperature difference, is taken as the input of BP network, and the load difference is used as the output. The neural network is trained to forecast the load difference. The sum of the load difference and the load of the reference day is the forecasted load. The method combines the advantages of the grey model and the neural network, and the validity is verified by simulation results.
出处
《现代电力》
2009年第2期1-4,共4页
Modern Electric Power
关键词
短期负荷预测
灰色模型
神经网络
基准日
short-term load forecasting
grey model
artificial neural network （ANN）
reference day
