基于深度神经网络的短期电力负荷预测模型研究

短期电力负荷预测对维护电网的稳定运行具有重要意义。在短期电力负荷预测中,深度神经网络模型具有较强的预测能力。本文结合GRU深度神经网络模型以及CNN深度神经网络模型,通过引入注意力机制,构建短期电力负荷预测模型。与未引未引入注意力机制的模型相比,在MAE上降低了22.59,在RMSE上降低了24.5,在MAPE上降低了1.74。

Short term power load forecasting is of great significance for maintaining the stable operation of the power grid.In short-term power load forecasting,deep neural network models have strong predictive ability.This article combines the GRU deep neural network model and the CNN deep neural network model,and constructs a short-term power load prediction model by introducing attention mechanism.Compared with the model without introducing attention mechanism,it decreased by 22.59 on MAE,24.5 on RMSE,and 1.74 on MAPE.