基于深度学习的雾霾质量浓度预测研究

Research onprediction of haze mass concentration based on deep learning

为有效地预测雾霾污染程度的主要评价指标PM2.5质量浓度,文章使用Blending集成学习策略并行连接CNN与LSTM,并建立基于CNN-LSTM集成学习的PM2.5质量浓度预测模型。经过真实数据验证,该模型对PM2.5质量浓度预测具有有效性,且相较于串联CNN-LSTM预测模型具有优越性。

In order to effectively predict the PM2.5 mass concentration of the main evaluation index of smog pollution degree,the paper uses Blending ensemble learning strategy to connect CNN and LSTM in parallel,and establish a PM2.5 mass concentration prediction model based on CNN-LSTM ensemble learning.After verification by real data,the model is effective for PM2.5 mass concentration prediction and is superior to the tandem CNN-LSTM prediction model.