数字农科院2.0

Improved PM2.5 prediction with spatio-temporal feature extraction and chemical components: The RCG-attention model

文献类型： 外文期刊

作者： Ao Li;Yafei Wang;Qianqian Qi;Yunfeng Li;Haixia Jia;Xin Zhou;Haixin Guo;Shuyang Xie;Junfeng Liu;Yujing Mu

作者机构：

关键词： Attention mechanism;Deep Learning;PM2.5;Prediction;Spatial and temporal feature

期刊名称： Science of the Total Environment

ISSN： 0048-9697

年卷期： 2024 年 955 卷

页码：

收录情况： SCIE（2024版） ; ; EI（2024版）

摘要： Deep learning models are widely used for PM2.5 prediction. However, neglecting temporal and spatial characteristics leads to low prediction accuracy. In this work, a new deep learning model (RCG - Attention model) was developed, which combines the residual neural network (ResNet) and the convolution gated recurrent network (ConvGRU) and is applied to extract the spatio - temporal features for predicting PM2.5 concentration over the subsequent 24 h. The ResNet extracts the spatial distribution features of pollutants, and the ConvGRU extracts temporal features. The spatial and temporal features are fused by the multi - head attention mechanism to obtain multi - dimensional features. These features are finally fed into a series of fully connected layers to predict the future results. Incorporating these chemical components enhances the scientific validity of the dataset and strengthens the inherent logical connections among variables. The Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and R - squared (R2) results indicate that the prediction performance of the RCG - Attention model surpasses that of other baseline models. The model demonstrates superior prediction performance across multiple monitoring stations, suggesting robust generalization capabilities and adaptability for various regions in one city. The SHAP results show that PM10, NO2, RH, NO3−, OC and NH4+ are significant influencing features. The RCG - Attention model provides a comprehensive solution for PM2.5 concentration prediction by integrating spatial and temporal feature extraction with chemical components.

分类号：

  • 相关文献

[1]A lightweight tea bud detection model based on Yolov5. Gui, Zhiyong,Chen, Jianneng,Li, Yang,Chen, Zhiwei,Wu, Chuanyu,Dong, Chunwang. 2023

[2]Automatic counting of rapeseed inflorescences using deep learning method and UAV RGB imagery. Li J.,Li Y.,Qiao J.,Li L.,Wang X.,Yao J.,Liao G.. 2023

[3]Semi-supervised learning and attention mechanism for weed detection in wheat. Teng Liu,Xiaojun Jin,Luyao Zhang,Jie Wang,Yong Chen,Chengsong Hu,Jialin Yu. 2023

[4]TBTA-D2Net: a novel hyperspectral image classification method based on triple-branch ternary-attention mechanism and Dense2Net. Ting Tang,Shengwei Zhang,Shaopeng Liu,Weihong Yan,Xin Pan. 2023

[5]Multi-Scale Mixed Attention Tea Shoot Instance Segmentation Model. Chen, Dongmei,Cao, Peipei,Yan, Lijie,Chen, Huidong,Lin, Jia,Li, Xin,Yuan, Lin,Wu, Kaihua. 2024

[6]An Unstructured Orchard Grape Detection Method Utilizing YOLOv5s. Wang, Wenhao,Shi, Yun,Liu, Wanfu,Che, Zijin. 2024

[7]MBSFC: hyperspectral image classification based on multi-branch and spectral feature conversion. Tang, Ting,Liu, Shaopeng,Fu, Xueliang,Yan, Weihong,Luo, Xiaoling,Pan, Xin. 2024

[8]Lightweight Structure and Attention Fusion for In-Field Crop Pest and Disease Detection. Zijing Luo,Yunsen Liang,Naimin Kong,Lirui Liang,Wenjun Peng,Yujie Yao,Chi Qin,Xiaohan Lu,Mingman Xu,Yining Zhang,Chenyang Lin,Chengyao Jiang,Mengyao Li,Yangxia Zheng,Yameng Jiang,Wei Lu. 2025

[9]ResNetKhib: a novel cell type-specific tool for predicting lysine 2-hydroxyisobutylation sites via transfer learning. Xiaoti Jia,Pei Zhao,Fuyi Li,Zhaohui Qin,Haoran Ren,Junzhou Li,Chunbo Miao,Quanzhi Zhao,Tatsuya Akutsu,Gensheng Dou,Zhen Chen,Jiangning Song. 2023

[10]大气环境对育肥猪舍内颗粒物浓度的影响. 刘杨,董红敏,陶秀萍,陈永杏,尚斌,徐文倩. 2016

[11]层叠笼密闭蛋鸡舍NH_3、CO_2、PMPM2.5、PMM10浓度变化及其影响因素分析. 沈丰菊,郑长山,李茜,张克强. 2017

[12]新乡冬季PM2.5中金属元素与水溶性离子年际变化及其来源解析. 闫广轩,雷豪杰,张靖雯,唐明双,张佳羽,曹治国,李云蓓,王跃思,樊静,李虎. 2019

[13]基于网络药理学、分子对接及实验验证探讨胖大海抗PM2.5诱导急性肺损伤的作用机制. 张凡,杜一凡,邓肖舒,张祖烽,韩先磊,田薇,李秀梅,陈勉,刘飞,王楠. 2025

[14]Comparison of Bacterial Community Structure in PM2.5 within Broiler Houses under Different Rearing Systems in China. Imran Khan,Wenxing Wang,Xiaobin Ye,Adamu Mani Isa,Muhammad Tariq Khan,Renna Sa,Lei Liu,Teng Ma,Hongfu Zhang. 2022

[15]Can Pm2.5 Pollution Worsen The Death Rate Due To Covid-19 In India And Pakistan?. Mehmood, K, Saifullah, Abrar, MM, Iqbal, M, Haider, E, Shoukat, HMH. 2020

[16]Bacterial Community Characteristics Shaped by Artificial Environmental PM2.5 Control in Intensive Broiler Houses. Wenxing Wang,Guoqi Dang,Imran Khan,Xiaobin Ye,Lei Liu,Ruqing Zhong,Liang Chen,Teng Ma,Hongfu Zhang. 2022

[17]Network pharmacology and phytochemical composition combined with validation in vivo and in vitro reveal the mechanism of platycodonis radix ameliorating PM2.5-induced acute lung injury. Han, Xianlei,Zhang, Yue,Zhang, Fan,Li, Xiumei,Meng, Yanli,Huo, Jinhai,Chen, Mian,Liu, Fei,Wang, Weiming,Wang, Nan. 2024

[18]Analysis of the Morphological Characteristics of PM2.5 and Its Microbiological Composition in a Fattening Pig House. Mingfeng Tang,Yue Jian,Jiaming Zhu,Kun Tian,Qiong Tan,Run Zhao. 2024

[19]Identification of Pine Wilt Disease Infected Wood Using UAV RGB Imagery and Improved YOLOv5 Models Integrated with Attention Mechanisms. Peng Zhang,Zhichao Wang,Yuan Rao,Jun Zheng,Ning Zhang,Degao Wang,Jianqiao Zhu,Yifan Fang,Xiang Gao. 2023

[20]Noise-tolerant RGB-D feature fusion network for outdoor fruit detection. Qixin Sun,Xiujuan Chai,Zhikang Zeng,Guomin Zhou,Tan Sun. 2022

作者其他论文 更多>>