基于自注意力机制的PM2.5长时间尺度预测
Long-term PM2.5 Concentrations Forecasting Based on Self-attention
何宇涵1
作者信息
- 1. 江西理工大学土木与测绘工程学院,江西赣州 341400
- 折叠
摘要
近些年基于机器学习的PM2.5预测逐渐成为主流,具有较强的非线性建模能力和提高预测精度的优势.然而,时间跨度较大的PM2.5浓度变化预测仍然面临挑战.文章构建了多种自注意力机制的模型,将PM2.5浓度的逐日预测提升到了14天的尺度,提升了以日为单位的PM2.5预测精度.并对 Informer、Auto-former、FEDformer和TCN模型在以日为单位的长时间尺度预测进行了对比分析,提高了PM2.5预测模型的准确性和可靠性.文章共构建了3,7,14天三个时间尺度,在各个时间尺度上,Autoformer模型性能表现都是最好的.相较于TCN模型,Autoformer在预测未来3天的时间尺度上,RMSE优化了43.36％,MAE优化了42.70％.在7天的时间尺度上,RMSE 优化了39.07％,MAE优化了8.98％、在14天的时间尺度上,RMSE优化了39.07％,MAE优化了8.98％.有效提升了PM2.5在长时间序列预测上的精度.
Abstract
In recent years,machine learning-based PM2.5 prediction has gradually become main-stream,with strong nonlinear modeling capabilities and advantages in improving prediction accu-racy.However,predicting PM2.5 concentration changes over a large time span still faces challen-ges.This study constructed multiple models with self-attention mechanisms,extending the daily prediction of PM2.5 concentration to a scale of 14 days and improving the prediction accuracy of PM2.5 on a daily basis.Comparative analysis of the Informer,Autoformer,FEDformer,and TCN models was conducted for long-term daily PM2.5 concentration prediction,enhancing the accuracy and reliability of PM2.5 prediction models.Three time scales were constructed in this study:3 days,7 days,and 14 days.The Auto former model performed the best at each time scale.Compared to the TCN model,the Autoformcr model showed significant performance improve-ments in predicting the future 3-day time scale,with RMSE and MAE reductions of 43.36％ and 42.70％,respectively.At the 7-day time scale,RMSE improved by 39.07％,and MAE im-proved by 8.98％.At the 14-day time scale,RMSE improved by 39.07％,and MAE improved by 8.98％.This study effectively improved the accuracy of PM2.5 prediction in long-term time scries forecasting.
关键词
PM2.5/长时间序列预测/自注意力机制/Autoformer/TCNKey words
PM2.5/LSTF/self-attention/Autoformer引用本文复制引用
出版年
2024
NETL
NSTL
万方数据