Computer Science ›› 2020, Vol. 47 ›› Issue (11A): 535-540.doi: 10.11896/jsjkx.200700164

• Big Data & Data Science • Previous Articles     Next Articles

Multi-scale Convolutional Neural Network Air Quality Prediction Model Based on Spatio-Temporal Optimization

ZHOU Jie1, LUO Yun-fang1, LEI Yao-jian2, LI Wen-jing3, FENG Yu1   

  1. 1 College of Mechanical,Electrical and Information Engineering,Guangxi Vocational & Technical College,Nanning 530226,China
    2 Guangxi College for Preshcool Education,Nanning 530022,China
    3 School of Logistics Management and Engineering,Nanning Normal University,Nanning 530001,China
  • Online:2020-11-15 Published:2020-11-17
  • About author:ZHOU Jie,born in 1992,postgraduate.His main research interests include intelligent computing and parallel computing.
    LUO Yun-fang,born in 1981,associate professor.His main research interests include big data,computer application technology and computer teaching.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China(61866006),Natural Science Foundation of Guangxi Education Department(2019ITA01002,2018KY0951,2019KY1220,2017KY0980) and Natural Science Project of Guangxi Vocational & Technical College(2019-176-191202).

Abstract: At present,the air quality prediction is mainly based on the time series of a single station,without considering the influence of the spatial characteristics on the air quality.To solve this problem,a multi-scale neural network (MSCNN-GALSTM) model based on spatiotemporal optimization is proposed for air quality prediction.one-dimensional multi-scale convolution kernel (MSCNN) is used to extract the local temporal and spatial characteristic relations in air quality data,the LINEAR SPLICING and fusion are carried out to obtain the space-time characteristic relation of multi-sites and multi-features,combine the advantage of long-short memory network (LSTM) to process time series,and introduce genetic algorithm (Ga) to optimize the parameter set of LSTM network globally,the time-space relationship of multi-site and multi-feature is input into the LSTM network,and then the long-term feature dependence of multi-site and multi-feature is output.Finally,the MSCNN-GALSTM model was compared with the single LSTM reference model and the single scale convolutional neural network model.The root mean square error (RMSE) decreased by about 11% and the average prediction accuracy increased by about 20%.The results show that the MSCNN-GALSTM model has more comprehensive feature extraction,deeper level,higher prediction accuracy and better generalization ability.

Key words: LSTM, Multiscale convolution, Prediction model, Spatial features, Spatio-temporal optimization

CLC Number: 

  • TP183
[1] ÖZELKADILAR G,KADILAR C.Assessing air quality in Aksaray with time series analysis[C]//American Institute of Physics Conference Series.AIP Publishing LLC,2017.
[2] RAHMAN N H A,LEE M H,SUHARTONO,et al.Artificial neural networks and fuzzy time series forecasting:an application to air quality[J].Quality & Quantity,2015,49(6):2633-2647.
[3] BIANCOFIORE F,BUSILACCHIO M,VERDECCHIA M,et al.Recursive neural network model for analysis and forecast of PM10 and PM2.5[J].Atmospheric Pollution Research,2017,8(4):652-659.
[4] SONG X,HUANG J,SONG D.Air Quality Prediction based on LSTM-Kalman Model[C]//2019 IEEE 8th Joint International Information Technology and Artificial Intelligence Conference (ITAIC).IEEE,2019:695-699.
[5] CHIOU-JYE H,PING-HUAN K.A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities[J].Sensors,2018,18(7):2220-2241.
[6] VERMA I,AHUJA R,MEISHERI H,et al.Air Pollutant Severity Prediction Using Bi-Directional LSTM Network[C]//2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI).ACM,2018:651-654.
[7] BECKERMAN B S,JERRETT M,SERRE M,et al.A hybridapproach to estimating national scale spatiotemporal variability of PM2.5 in the contiguous United States[J].Environmental Science & Technology,2013,47(13):7233-7241.
[8] HWA-LUNG Y,CHIH-HSIN W.Retrospective prediction ofintraurban spatiotemporal distribution of PM2.5 in Taipei[J].Atmospheric Environment,2010,44(25):3053-3065.
[9] BARI M A,KINDZIERSKI W B.Characteristics of air quality and sources affecting fine particulate matter (PM2.5) levels in the City of Red Deer,Canada[J].Environmental Pollution,2016,221:367-376.
[10] LIAO Z H,SUN J R,FAN S J,et al.Variation characteristics and influencing factors of air pollution in Pearl River Delta area from 2006 to 2012[J].ZhongguoHuanjingKexue/china Environmental Science,2015,35(2):329-336.
[11] LI X,PENG L,YAO X,et al.Long short-term memory neural network for air pollutant concentration predictions:Method development and evaluation[J].Environmental Pollution,2017,231(1):997-1004.
[12] ZHU Y J,LI Q,HOU J X,et al.Spatio-temporal modelingand prediction of PM2.5concentration based on Bayesian method[J].Science of Surveying and Mapping,2016,41(2):44-48.
[13] HAN X G,LI B Y,GUAN Z Y.Atmospheric Quality Prediction Model Based on RBF Neural Network-Markov Chain of Grey Correlational Analysis Filter Indexes[J].Acta ScientiarumNaturalium Universitatis Nankaiensis,2013,46(2):22-27.
[14] GAO S,HU H P,LI Y,et al.AQI Prediction Based on Improved Mind EvolutionaryAlgorithm and BP Neural Network[J].Mathematics in Practice and RACTICE Theory,2018,48(19):151-157.
[15] YUAN G,YANG W.Evaluating China's Air Pollution Control Policy with Extended AQI Indicator System:Example of the Beijing-Tianjin-Hebei Region[J].Sustainability,2019,11(3):939.
[16] HU Z,LI W,QIAO J.Prediction of PM2.5 based on Elman neural network with chaos theory[C]//2016 35th Chinese Control Conference (CCC).IEEE,2016:3573-3578.
[17] SAYEGH A S,MUNIR S,HABEEBULLAH T M.Comparing the Performance of Statistical Models for Predicting PM10 Concentrations[J].Aerosol & Air Quality Research,2014,14(3):653-665.
[18] GLOROT X,BENGIO Y.Understanding the difficulty of training deep feedforward neural networks[J].Journal of Machine Learning Research,2010(9):249-256.
[19] YE X Z,TAO F F,QI R Z,et al.Improvement on Activation Functions of Recurrent Neural Network Architectures[J].Jisuanji Yu Xiandaihua,2016(12):29-33.
[20] LI H D.Portforlio selection based on recurrent neural network[D].Zhengzhou:Zhengzhou University,2018.
[21] DING L,FANG W,LUO H,et al.A deep hybrid learning model to detect unsafe behavior:Integrating convolution neural networks and long short-term memory[J].Automation in Construction,2018,86:118-124.
[22] LEI Y J.Research on Urban Air Quality Prediction Based onTemporal and Spatial Optimization Neural Network[D].Nanning:Nanning Normal University,2020.
[23] JI L.Research and Implementation ofPM2.5 Prediction Based on CNNs-GRU Deep Learning[D].Chongqing:Chongqing University of Posts and Telecommunications,2019.
[1] ZHANG Yuan, KANG Le, GONG Zhao-hui, ZHANG Zhi-hong. Related Transaction Behavior Detection in Futures Market Based on Bi-LSTM [J]. Computer Science, 2022, 49(7): 31-39.
[2] JIN Fang-yan, WANG Xiu-li. Implicit Causality Extraction of Financial Events Integrating RACNN and BiLSTM [J]. Computer Science, 2022, 49(7): 179-186.
[3] LIN Xi, CHEN Zi-zhuo, WANG Zhong-qing. Aspect-level Sentiment Classification Based on Imbalanced Data and Ensemble Learning [J]. Computer Science, 2022, 49(6A): 144-149.
[4] WANG Shan, XU Chu-yi, SHI Chun-xiang, ZHANG Ying. Study on Cloud Classification Method of Satellite Cloud Images Based on CNN-LSTM [J]. Computer Science, 2022, 49(6A): 675-679.
[5] YAN Rui, LIANG Zhi-yong, LI Jin-tao, REN Fei. Predicting Tumor-related Indicators Based on Deep Learning and H&E Stained Pathological Images:A Survey [J]. Computer Science, 2022, 49(2): 69-82.
[6] DING Feng, SUN Xiao. Negative-emotion Opinion Target Extraction Based on Attention and BiLSTM-CRF [J]. Computer Science, 2022, 49(2): 223-230.
[7] YUAN Jing-ling, DING Yuan-yuan, SHENG De-ming, LI Lin. Image-Text Sentiment Analysis Model Based on Visual Aspect Attention [J]. Computer Science, 2022, 49(1): 219-224.
[8] CHENG Si-wei, GE Wei-yi, WANG Yu, XU Jian. BGCN:Trigger Detection Based on BERT and Graph Convolution Network [J]. Computer Science, 2021, 48(7): 292-298.
[9] XING Hao, LI Ming. Deepfake Video Detection Based on 3D Convolutional Neural Networks [J]. Computer Science, 2021, 48(7): 86-92.
[10] HU Yu-wen. Stock Forecast Based on Optimized LSTM Model [J]. Computer Science, 2021, 48(6A): 151-157.
[11] CHEN Hui-qin, GUO Guan-cheng, QIN Chao-xuan, LI Zhao-bi. Research on Elderly Population Prediction Based on GM-LSTM Model in Nanjing City [J]. Computer Science, 2021, 48(6A): 231-234.
[12] YU Jian-ye, QI Yong, WANG Bao-zhuo. Distributed Combination Deep Learning Intrusion Detection Method for Internet of Vehicles Based on Spark [J]. Computer Science, 2021, 48(6A): 518-523.
[13] ZHANG Zheng-wan, WU Di, ZHANG Chun-jiong. Study of Cellular Traffic Prediction Based on Multi-channel Sparse LSTM [J]. Computer Science, 2021, 48(6): 296-300.
[14] DONG Zhe, SHAO Ruo-qi, CHEN Yu-liang, ZHAI Wei-feng. Named Entity Recognition in Food Field Based on BERT and Adversarial Training [J]. Computer Science, 2021, 48(5): 247-253.
[15] LI Bing-rong, PI De-chang, HOU Meng-ru. Destination Prediction of Moving Objects Based on Convolutional Neural Networks and Long-Short Term Memory [J]. Computer Science, 2021, 48(4): 70-77.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!