基于双向长短期记忆网络的区域电网新能源消纳预测算法

Prediction algorithm of new energy consumption in regional power grid based on bidirectional long short term memory network

目前,区域电网新能源消纳问题日益严重,采用优化算法进行区域电网新能源消纳评估的技术难度大、求解效率低、消耗时间长。为此,本文提出基于双向长短期记忆网络(BiLSTM)的区域电网新能源消纳预测算法。首先分析影响区域电网新能源消纳的因素,并进行新能源消纳预测数据准备;然后提出基于BiLSTM的区域电网新能源消纳预测算法,利用电网历史运行数据训练模型,实现区域电网新能源消纳的快速准确在线预测;最后利用实际电网数据验证了所提算法的有效性,为电网运行人员提供参考。

In view of the increasingly serious problem of new energy consumption in regional power grids,and the technical difficulty,low efficiency and long consumption time of using optimization algorithms to evaluate the new energy consumption in regional power grids,this paper proposes a new energy consumption prediction method based on bidirectional long short term memory network(BiLSTM)for regional power grids.Firstly,the factors that affect the new energy consumption of regional power grid are analyzed,and the new energy consumption forecast data is prepared.Then,a new energy consumption prediction algorithm for regional power grids based on BiLSTM is proposed,and the historical power grid operation data training model is used to realize fast and accurate online prediction of new energy consumption in regional power grids.Finally,the validity of the method proposed in this paper is verified by actual power grid data,which provides reference for power grid operators.