利用2DGRA-BiLSTM模型的日前光伏功率曲线预测方法

Day-ahead photovoltaic power curve predictionmethod using 2DGRA-BiLSTM model

为了克服光伏发电固有的间断性和波动性对电网稳定性的负面影响,提出一种二维灰度关联分析-双向长短期记忆神经网络(two-dimensional grey relational analysis and bidirectional long short-term memory network, 2DGRA-BiLSTM)模型,用于实现日前光伏功率曲线预测,以更好指导电网调度.不同于以往的点预测,本研究将日功率曲线作为整体进行预测.首先用2DGRA实现最佳历史相似日数据的获取;其次,根据日功率曲线的波动性将总数据分为3类;最后,根据3种分类,分别训练3种BiLSTM模型对日功率曲线进行预测.所提出的预测模型通过沙漠知识澳大利亚太阳能中心历史气象和功率数据进行训练,并通过数值天气预报和功率数据进行测试.对比其他几种神经网络模型,实验表明所提出模型具有更好的综合预测性能,在晴空、轻度非晴空和重度非晴空条件下,决定系数(R2)分别为0.994、0.940和0.782.

In order to overcome the negative impact on the grid caused by photovoltaic(PV)power generation,a 2DGRA-BiLSTM(two-dimensional grey relational analysis and bidirectional long short-term memory network)model is proposed for day-ahead PV power curve prediction.Firstly,2DGRA was used to obtain best historical similar day.Secondly,the total data is divided into three classifications according to the fluctuation of the daily power curve.Finally,three BiLSTM models were trained to predict the daily power curves.The proposed model was trained with historical meteorological and power data from Desert Knowledge Australia Solar Centre(DKASC),and tested with numerical weather prediction(NWP)and power data.Compared with other neural network models,the experimental results show that the proposed model has better comprehensive prediction performance,with coefficient of determination(R 2)value of 0.994,0.940 and 0.782 in clear sky,mild non clear sky and severe non clear sky condition,respectively.