基于XGBoost-LSTM组合模型的光伏发电功率预测

PHOTOVOLTAIC POWER PREDICTION BASED ON COMBINED XGBOOST-LSTM MODEL

该文提出一种基于极端梯度提升(XGBoost)模型和长短期记忆网络(LSTM)模型的短期光伏发电功率预测组合模型。根据短期光伏发电特性,首先分别建立XGBoost模型和LSTM模型,然后利用XGBoost模型进行初步预测增加特征,并利用误差倒数法将两模型组合起来进行预测。选取2018年光伏电站人工智能运维大数据处理分析大赛的数据集进行实验评估,最终结果表明,该文所构建的XGBoost-LSTM组合模型的均方根误差(RMSE)为0.214,将上述方法与随机森林、GBDT模型和单一的XGBoost模型和LSTM模型相比较,该文提出的方法具有更高的预测精度。

A combination model based on XGBoost(eXtreme Gradient Boosting)model and LSTM(Long Short Term Memory)model is proposed in this paper.According to the characteristics of short-term photovoltaic power generation,the XGBoost model and the LSTM model are established respectively in the first place.Then,the XGBoost model is used for preliminary prediction to add features,and the error reciprocal method is used to combine the two models for prediction.Data sets from 2018 big data processing and analysis contest of photovoltaic power station artificial intelligence operation and maintenance are selected for experimental evaluation.The final result shows that the root-mean-square error(RMSE)of the constructed XGBoost-LSTM combination model is 0.214.Compared with the Random Forest,GBDT model,XGBoost model,LSTM model,the proposed method has higher prediction accuracy.