基于LSTM网络的时间多尺度电采暖负荷预测

电采暖负荷的准确预测对配电网安全稳定运行具有重要作用。为提高电采暖负荷预测精度,提出基于长短期记忆网络的时间多尺度电采暖负荷预测方法。该方法首先确定负荷预测网络模型参数,通过分析不同时间尺度下的电采暖负荷数据特征,在不同训练步长下,建立基于长短期记忆网络的电采暖负荷预测模型,实现电采暖负荷的准确预测。采用北京某地的电采暖负荷实测数据进行验证,结果表明在不同的时间尺度下,长短期记忆网络均能够实现电采暖负荷的准确预测。

Multi-time Scale Electric Heating Load Forecasting Based on Long Short-term Memory Network

The accurate forecasting of electric heating load plays an important role in the safe and stable operation of dis?tribution network.To improve the forecasting accuracy,a long short-term memory(LSTM)network based multi timescale electric heating load forecasting method is proposed in this paper.First,the parameters of the load forecasting net?work model are determined.Then,by analyzing the characteristics of electric heating load data at different time scales,an electric heating load forecasting model based on the LSTM network is established to realize an accurate prediction of electric heating load.The measurement data of electric heating load from one region in Beijing are used to verify the pro?posed forecasting method,indicating that the LSTM network can accurately predict the electric heating load at different time scales.