基于一种样本卷积和交互网络模型的空气质量预测

覃业梅; 胡博飓; 冯懿归; 周帆; 赵慎

doi:10.11959/j.issn.2096-6652.202428

您当前的位置：

首页 >

文章列表页 >

基于一种样本卷积和交互网络模型的空气质量预测

学术论文 | 更新时间：2024-11-04

- 基于一种样本卷积和交互网络模型的空气质量预测
- Air quality forecasting based on a sample convolution and interaction network model
- 智能科学与技术学报 2024年6卷第3期页码：356-366
- 作者机构：
  
  1.湖南工商大学智能工程与智能制造学院，湖南长沙 410205
  2.湘江实验室，湖南长沙 410205
  3.国防科技大学计算机学院，湖南长沙 410073
- 作者简介：
  
  [ "覃业梅（1979- ），女，博士，湖南工商大学智能工程与智能制造学院副教授，电子信息工程系主任，主要研究方向为时间序列信号分析与预测。" ]
  [ "胡博飓（1999- ），男，湖南工商大学智能工程与智能制造学院硕士生，主要研究方向为时间序列预测。" ]
  [ "冯懿归（1997- ），男，国防科技大学计算机学院博士生，主要研究方向为时间序列预测、大模型。" ]
  [ "周帆（2001- ），女，湖南工商大学智能工程与智能制造学院硕士生，主要研究方向为时间序列预测。" ]
  [ "赵慎（1983- ），男，博士，湖南工商大学智能工程与智能制造学院副教授，智能科学与技术系主任，主要研究方向为阵列信号处理。" ]
- 基金信息：
  
  湖南省教育厅科学研究重点项目(21A0381;23A0464);湘江实验室重大项目(22XJ01002;23XJ02006)
- DOI：10.11959/j.issn.2096-6652.202428
  中图分类号： TP183;X831
- 收稿：2024-03-08，
  
  修回：2024-08-12，
  
  纸质出版：2024-09-15
- 稿件说明：
移动端阅览
覃业梅,胡博飓,冯懿归等.基于一种样本卷积和交互网络模型的空气质量预测[J].智能科学与技术学报,2024,06(03):356-366.

QIN Yemei,HU Boju,FENG Yigui,et al.Air quality forecasting based on a sample convolution and interaction network model[J].Chinese Journal of Intelligent Science and Technology,2024,06(03):356-366.
覃业梅,胡博飓,冯懿归等.基于一种样本卷积和交互网络模型的空气质量预测[J].智能科学与技术学报,2024,06(03):356-366. DOI： 10.11959/j.issn.2096-6652.202428.

QIN Yemei,HU Boju,FENG Yigui,et al.Air quality forecasting based on a sample convolution and interaction network model[J].Chinese Journal of Intelligent Science and Technology,2024,06(03):356-366. DOI： 10.11959/j.issn.2096-6652.202428.

摘要

空气质量预测是治理和减轻空气污染的有效手段。为了提高预测精度，构建了一个新的空气质量预测模型，即样本卷积和交互网络（sample convolutional and interaction network，SCINet）模型。该模型由多个SCI-Block按照完全二叉树结构排列而成，通过翻转奇偶分裂重新排列生成一个新的序列，该结构能够更好地捕捉多变量大气污染物彼此间复杂的依赖关系和局部趋势。因为大气污染物监测数据具有季节性和随机性，所以使用两个SCINet进行叠加，既能扩大卷积运算的接受域，又能实现多分辨率分析。此外，通过模型深度及超参数调优，使其更加拟合空气质量时序数据特征，能够有助于提取目标变量的时间关系特征。最后，通过北京PM

2.5

数据集和北京多站点空气质量数据集进行实证研究，结果表明，SCINet模型具有更高的预测精度，在短期预测中其均方根误差比对比模型中表现最佳的DAQFF模型减少了31.59%，在长期预测中减少了24.36%。

Abstract

Air quality forecasting is an effective means of managing and mitigating air pollution. To enhance predicti

on accuracy

a new air quality prediction model

namely the sample convolutional and interaction network(SCINet)

is introduced in this paper. The model is composed of multiple SCI-Blocks arranged in a complete binary tree structure. Whereafter

the time series is rearranged through flipping odd-even splits

and a new sequence is generated

which is able to capture the complex dependencies and local trends of multivariate atmospheric pollutants better. Given the seasonality and randomness of monitoring data for atmospheric pollutant

the paper employs two SCINets for stacking

which not only expands the receptive field of convolutional operations

but also enables multi-resolution analysis. Furthermore

through the optimization of model depth and hyperparameters

the model may fit the temporal characteristics of air quality time series data better

which is helpful to extract the temporal relationship features of the target variable. In the end

the Beijing PM

2.5

dataset and the Beijing multi-site air quality dataset are utilized to evaluate the effectiveness of SCINet. The results show that SCINet has higher prediction accuracy

whose the root mean square error (

RMSE

) is reduced by 31.59% in short-term prediction and 24.36% in long-term prediction compared with the best-performing DAQFF model.

关键词

Keywords

references

ZHANG Y, BOCQUET M, MALLET V, et al. Real-time air quality forecasting, part I: history, techniques, and current status[J]. Atmospheric Environment, 2012, 60: 632-655.

ZHANG Y, BOCQUET M, MALLET V, et al. Real-time air quality forecasting, part II: state of the science, current research needs, and future prospects[J]. Atmospheric Environment, 2012, 60: 656-676.

吴钲, 谢旻, 高阳华, 等. 利用集合均方根卡尔曼滤波反演重庆地区SO 2 源排放[J ] . 环境科学研究, 2018, 31(1): 25-33.

WU Z, XIE M, GAO Y H, et al. Inversion of SO 2 emissions over Chongqing with ensemble square root Kalman filter[J ] . Research of Environmental Sciences, 2018, 31(1): 25-33.

LIU C F, WU C Y, KANG X Y, et al. Evaluation of the prediction performance of air quality numerical forecast models in Shenzhen[J]. Atmospheric Environment, 2023, 314: 120058.

WANG M L, ZHANG Y M, FUNG J C H, et al. A novel hybrid clustering model of region segmentation to fuse CMAQ simulations with observations[J]. Atmospheric Environment, 2022, 278: 119062.

王玲, 康子豪. 基于时空组Lasso与分层贝叶斯时空模型的变量选择方法[J]. 地球信息科学学报, 2023, 25(7): 1312-1324.

WANG L, KANG Z H. Variable selection method based on spatio-temporal group lasso and hierarchical Bayesian spatio-temporal model[J]. Journal of Geo-Information Science, 2023, 25(7): 1312-1324.

郭利进, 井海明, 南亚翔, 等. 基于卡尔曼滤波融合算法的空气质量指数预测[J]. 环境污染与防治, 2017, 39(4): 388-391.

GUO L J, JING H M, NAN Y X, et al. Prediction of air quality index based on Kalman filtering fusion algorithm[J]. Environmental Pollution & Control, 2017, 39(4): 388-391.

MANI G, VISWANADHAPALLI J K, ALEXANDER S A, et al. Prediction and forecasting of air quality index in Chennai using regression and ARIMA time series models[J]. Journal of Engineering Research, 2021, 9: 179-194.

SHARMA E, DEO R C, PRASAD R, et al. A hybrid air quality early-warning framework: an hourly forecasting model with online sequential extreme learning machines and empirical mode decomposition algorithms[J]. The Science of the Total Environment, 2020, 709: 135934.

KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.

KARPATHY A, LI F F. Deep visual-semantic alignments for generating image descriptions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2015: 3128-3137.

CHO K, VAN MERRIENBOER B, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[EB ] . arXiv preprint, 2014, arXiv: 1406.1078 .

LIPTON Z C, BERKOWITZ J, ELKAN C. A critical review of recurrent neural networks for sequence learning[EB ] . arXiv preprint, 2015, arXiv: 1506.00019 .

LÄNGKVIST M, KARLSSON L, LOUTFI A. A review of unsupervised feature learning and deep learning for time-series modeling[J]. Pattern Recognition Letters, 2014, 42: 11-24.

YANG J B, NGUYEN M N, SAN P P, et al. Deep convolutional neural networks on multichannel time series for human activity recognition[J]. IJCAI International Joint Conference on Artificial Intelligence, 2015(1): 3995-4001.

GAMBOA J C B. Deep learning for time-series analysis[EB ] . arXiv preprint, 2017, arXiv: 1701.01887 .

QI Z A, WANG T C, SONG G J, et al. Deep air learning: interpolation, prediction, and feature analysis of fine-grained air quality[J]. IEEE Transactions on Knowledge and Data Engineering, 2018, 30(12): 2285-2297.

ERAVCI B, FERHATOSMANOGLU H. Diverse relevance feedback for time series with autoencoder based summarizations[J]. IEEE Transactions on Knowledge and Data Engineering, 2018, 30(12): 2298-2311.

WANG J S, SONG G J. A deep spatial-temporal ensemble model for air quality prediction[J]. Neurocomputing, 2018, 314: 198-206.

ZHOU H Y, ZHANG S H, PENG J Q, et al. Informer: beyond efficient transformer for long sequence time-series forecasting[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(12): 11106-11115.

DENG A L, HOOI B. Graph neural network-based anomaly detection in multivariate time series[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(5): 4027-4035.

李明明, 王小兰, 岳江, 等. 融合EOF分解及CNN-LSTM神经网络的PM 2.5 预测[J ] . 环境科学, 2024: 1-16.

LI M M, WANG X L, YUE J, et al. PM 2.5 prediction based on fusion of EOF decomposition and CNN-LSTM neural network [J ] . Environmental Science, 2024: 1-16.

CHEN J, YUAN C W, DONG S, et al. A novel spatiotemporal multigraph convolutional network for air pollution prediction[J]. Applied Intelligence, 2023, 53(15): 18319-18332.

GILIK A, OGRENCI A S, OZMEN A. Air quality prediction using CNN+LSTM-based hybrid deep learning architecture[J]. Environmental Science and Pollution Research, 2022, 29(8): 11920-11938.

廖海斌, 袁理, 龚颢巍. 基于多尺度时空图神经网络的污染物浓度预测[J]. 控制与决策, 2024, 39(4): 1396-1402.

LIAO H B, YUAN L, GONG H W. Pollutant concentration forecast based on multiscale spatiotemporal graph neural network[J]. Control and Decision, 2024, 39(4): 1396-1402.

LIU M H, ZENG A L, CHEN M X, et al. SCINet: time series modeling and forecasting with sample convolution and interaction[EB ] . arXiv preprint, 2021, arXiv: 2106.09305 .

ZHENG Y, LIU F R, HSIEH H P. U-Air: when urban air quality inference meets big data[C]//Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining. New York: ACM, 2013: 1436-1444.

ZHENG Y, YI X W, LI M, et al. Forecasting fine-grained air quality based on big data[C]//Proceedings of the Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2015: 2267-2276.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于高维特征表示的交通场景识别

Cerberus：基于深度学习的跨站社交网络机器人检测系统