您的位置：山东大学 -> 科技期刊社 -> 《山东大学学报（工学版）》

山东大学学报(工学版) ›› 2015, Vol. 45 ›› Issue (6): 76-83.doi: 10.6040/j.issn.1672-3961.0.2014.214

• 土木工程 • 上一篇    下一篇

基于气象因素的PM2.5质量浓度预测模型

刘杰1, 杨鹏2, 吕文生1, 刘阿古达木1, 刘俊秀2   

  1. 1. 北京科技大学土木与环境工程学院, 北京 100083;
    2. 北京市信息服务工程重点实验室(北京联合大学), 北京 100101
  • 收稿日期:2014-08-05 修回日期:2015-06-02 出版日期:2015-12-20 发布日期:2014-08-05
  • 通讯作者: 杨鹏(1965-),男,四川大英人,教授,博士,博导,主要研究方向为矿业工程,安全工程与系统工程研究.E-mail:yangpeng@buu.edu.cn E-mail:yangpeng@buu.edu.cn
  • 作者简介:刘杰(1986-),男,云南禄丰人,博士研究生,主要研究方向为城市空气质量监测,矿井通风与安全.E-mail:liujie19860809@qq.com
  • 基金资助:
    北京市属高等学校高层次人才引进与培养——"长城学者"培养计划资助项目(CIT&TCD20130320)

Prediction models of PM2.5 mass concentration based on meteorological factors

LIU Jie1, YANG Peng2, LYU Wensheng1, LIU Agudamu1, LIU Junxiu2   

  1. 1. School of Civil and Environmental Engineering, University of Science and Technology Beijing, Beijing 100083, China;
    2. Beijing Key Laboratory of Information Service Engineering, Beijing Union University, Beijing 100101, China
  • Received:2014-08-05 Revised:2015-06-02 Online:2015-12-20 Published:2014-08-05

PDF (PC)

1451

摘要： 为得出拟合效果最佳的预测模型,建立了多元回归和机器学习预测模型对PM2.5质量浓度进行预测。在输入气象因素的基础上,引入污染物质量浓度基础值和周期因素两类变量作为预测输入,并对4种预测模型进行对比研究。研究结果表明:对预测输入进行改进后,多元线性回归预测模型拟合优度由0.52提高至0.64,所选取的气象参数、污染物质量浓度基础值和周期因素能较好地描述PM2.5质量浓度的日变化情况;与多元线性回归预测模型相比,BP神经网络和支持向量机两种预测模型能较好地捕捉PM2.5质量浓度与预测输入之间的非线性影响规律,整体拟合优度分别达0.69和0.74,预测准确度较高;支持向量机预测模型可作为PM2.5质量浓度预测的首选方法。

关键词: 机器学习, BP神经网络, 支持向量机, 多元线性回归, PM2.5

Abstract: In order to get the optimal prediction model, the prediction models of PM2.5 mass concentration based on multiple linear regression and machine learning were developed. Basic values of pollutants mass concentrations and periodical factors were introduced as predictive inputs based on meteorological factors. Then four prediction models were developed for comparison. Results showed that goodness of fit of multiple linear regression model based on improved predictive inputs was increased from 0.52 to 0.64. The selected meteorological factors, basic values of pollutants mass concentrations and periodical factors could accurately describe daily variation of PM2.5. BP neural network and support vector machine models could be trained to model the highly non-linear relationships between PM2.5 mass concentration and predictive inputs. They provided satisfactory results with goodness of fit of 0.69 and 0.74, respectively. Support vector machine model was proved to be optimal prediction model of PM2.5 mass concentration.

Key words: PM2.5, multiple linear regression, support vector machine, machine learning, BP neural network

中图分类号: 

  • X831
[1] 刘杰, 杨鹏, 吕文生. 北京大气颗粒物污染特征及空间分布插值分析[J]. 北京科技大学学报, 2014, 36(9):1269-1279. LIU Jie, YANG Peng, L Wensheng. Pollution characteristics of particulate matters and interpolation analysis of their spatial distribution in Beijing[J]. Journal of University of Science and Technology Beijing, 2014, 36(9):1269-1279.
[2] 刘杰, 杨鹏, 吕文生, 等. 模糊时序与支持向量机建模相结合的PM2.5质量浓度预测[J]. 北京科技大学学报, 2014, 36(12):1694-1702. LIU Jie, YANG Peng, L Wensheng, et al. Prediction model of PM2.5 mass concentrations based on fuzzy time series and support vector machine[J]. Journal of University of Science and Technology Beijing, 2014, 36(12):1694-1702.
[3] ZIOMAS I C, MELAS D, ZEREFOS C S, et al. Forecasting peak pollutant levels from meteorological variables[J]. Atmospheric Environment, 1995, 29(24):3703-3711.
[4] CHALOULAKOU A, KASSOMENOS P, SPYRELLIS N, et al. Measurements of PM10 and PM2.5 particle concentrations in Athens, Greece[J]. Atmospheric Environment, 2003, 37(5):649-660.
[5] HUSSEIN T, KARPPINEN A, KUKKONEN J, et al. Meteorological dependence of size-fractionated number concentrations of urban aerosol particles[J]. Atmospheric Environment, 2006, 40(8):1427-1440.
[6] 于涛, 王可丽, 李芳, 等. 兰州市大气环境质量影响因素多元统计模型研究[J]. 西安建筑科技大学学报:自然科学版, 2009, 41(5):723-729. YU Tao, WANG Keli, LI Fang, et al. Study on multi-variate statistical analysis model of influencing factors on the atmospheric environmental quality in Lanzhou city[J]. Journal of Xi'an University of Archrchitecture and Technology:Natural Science Edition, 2009, 41(5):723-729.
[7] KUKKONEN J, PARTANEN L, KARPPINEN A, et al. Extensive evaluation of neural network models for the prediction of NO2 and PM10 concentrations, compared with a deterministic modelling system and measurement in central Helsinki[J]. Atmospheric Environment, 2003, 37(32):4539-4550.
[8] MCKENDRY I. Evaluation of artificial neural networks for fine particulate pollution(PM10 and PM2.5) forecasting[J]. Journal of the Air & Waste Management Association, 2002, 52(9):1096-1101.
[9] 石灵芝, 邓启红, 路婵, 等. 基于BP人工神经网络的大气颗粒物PM10质量浓度预测[J]. 中南大学学报:自然科学版, 2012, 43(5):1969-1974. SHI Lingzhi, DENG Qihong, LU Chan, et al. Prediction of PM10 mass concentrations based on BP artificial neural network[J]. Journal of Central University:Science and Technology, 2012, 43(5):1969-1974.
[10] GRIVAS G, CHALOULAKOU A. Artificial neural network models for prediction of PM10 hourly concentrations, in the Greater Area of Athens, Greece[J]. Atmospheric Environment, 2006, 40(7):1216-1229.
[11] CAI M, YIN Y, XIE M. Prediction of hourly air pollutant concentrations near urban arterials using artificial neural network approach[J]. Transportation Research:Part D, 2009, 14(1):32-41.
[12] 孟健, 付桦, 赵晓林, 等. 北京西三环路分时段车流量分析[J]. 首都师范大学学报:自然科学版, 2006, 27(2):89-98. MENG Jian, FU Hua, ZHAO Xiaolin, et al. Research progress on the mechanism of nitrogen and phosphorus in non-point source pollution[J]. Journal of Capital Normal University:Natural Science Edition, 2006, 27(2):89-98.
[13] 潘本锋, 赵熠琳, 李健军, 等. 气象因素对大气中PM2.5的去除效应[J]. 环境科技, 2012, 25(6):41-44. PAN Benfeng, ZHAO Yilin, LI Jianjun, et al. Analysis of the scavenging efficiency on PM2.5 concentration of some kinds of meteorological factors[J]. Environment Science and Technology, 2012, 25(6):41-44.
[14] CAPPARUCCIA R, DE L R, MARCHITTO E. Integrating support vector machines and neural networks[J]. Neural Networks, 2007, 20(5):590-597.
[15] WANG Wenjian, MEN CHANGqian, LU Weizhen. Online prediction model based on support vector machine[J]. Neurocomputing, 2008, 71(4/6):550-558.
[16] ANGUITA D, CHIO A, PISCHIUTTA S. A support vector machine with integer parameters[J]. Neurocomputing, 2008, 72(1/3):480-489.
[17] 姜雪, 卢文喜, 杨青春, 等. 应用支持向量机评价土壤环境质量[J]. 中国环境科学, 2014, 34(5):1229-1235. JIANG Xue, LU Wenxi, YANG Qingchun, et al. Application of support vector machine in soil environmental quality assessment[J]. China Environmental Science, 2014, 34(5):1229-1235.
[18] CORTES C, VAPNIK V. Support-Vector networks[J]. Machine Learning, 1995(20):273-297.
[19] ANDERSON J. An introduction to neural networks[M]. London, the UK:MIT Press, 1995.
[20] CANEVALE C, FINZI G, PISONI E, et al. Neuro-fuzzy and neural network systems for air quality control[J]. Atmospheric Environment, 2008, 7(64):1-11.
[21] SANCHEZ A D. Advanced support vector machines and kernel methods[J]. Neurocomputing, 2003, 55(1):5-20.
[22] 刘春波, 王群芳, 潘丰. 基于蚁群优化算法的支持向量机参数选择及仿真[J]. 中南大学学报:自然科学版, 2008, 39(6):1309-1313. LIU Chunbo, WANG Qunfang, PAN Feng. Parameters selection and stimulation of support vector machines based on ant colony optimization algorithm[J]. Journal of Central South University:Science and Technology, 2008, 39(6):1309-1313.
[23] RAKOTOMAMONJYA A, RICHE R L, GUALANDRISC D, et al. A comparison of statistical learning approaches for engine torque estimation[J]. Control Engineering Practice, 2008, 16(1):43-55.
[24] CHALOULAKOU A, GRIVAS G, SPYRELLIS N. Neural network and multiple regression models for PM10 prediction in Athens:a comparative assessment[J]. Journal of the Air & Waste Management Association, 2003, 53(10):1183-1190.
[25] LIU W Z, WANG W J, WANG X K, et al. Potential assessment of a neural network model with PCA/RBF approach for forecasting pollutant trends in Mong Kok urban air, Hong Kong[J]. Environment Research, 2004, 96(1):79-87.
[1] 张冕,黄颖,梅海艺,郭毓. 基于Kinect的配电作业机器人智能人机交互方法[J]. 山东大学学报(工学版), 2018, 48(5): 103-108.
[2] 叶明全,高凌云,万春圆. 基于人工蜂群和SVM的基因表达数据分类[J]. 山东大学学报(工学版), 2018, 48(3): 10-16.
[3] 韩学山,王俊雄,孙东磊,李文博,张心怡,韦志清. 计及空间关联冗余的节点负荷预测方法[J]. 山东大学学报(工学版), 2017, 47(6): 7-12.
[4] 刘洋,刘博,王峰. 基于Parameter Server框架的大数据挖掘优化算法[J]. 山东大学学报(工学版), 2017, 47(4): 1-6.
[5] 刘岩,李幼军,陈萌. 基于EMD和SVM的抑郁症静息态脑电信号分类研究[J]. 山东大学学报(工学版), 2017, 47(3): 21-26.
[6] 李素姝,王士同,李滔. 基于LS-SVM与模糊补准则的特征选择方法[J]. 山东大学学报(工学版), 2017, 47(3): 34-42.
[7] 魏波,张文生,李元香,夏学文,吕敬钦. 一种选择特征的稀疏在线学习算法[J]. 山东大学学报(工学版), 2017, 47(1): 22-27.
[8] 周旺,张晨麟,吴建鑫. 一种基于Hartigan-Wong和Lloyd的定性平衡聚类算法[J]. 山东大学学报(工学版), 2016, 46(5): 37-44.
[9] 孟令恒,丁世飞. 基于单静态图像的深度感知模型[J]. 山东大学学报(工学版), 2016, 46(3): 37-43.
[10] 刘晓勇. 一种基于树核函数的半监督关系抽取方法研究[J]. 山东大学学报(工学版), 2015, 45(2): 22-26.
[11] 浩庆波, 牟少敏, 尹传环, 昌腾腾, 崔文斌. 一种基于聚类的快速局部支持向量机算法[J]. 山东大学学报(工学版), 2015, 45(1): 13-18.
[12] 郑毅, 朱成璋. 基于深度信念网络的PM2.5预测[J]. 山东大学学报(工学版), 2014, 44(6): 19-25.
[13] 李发权, 杨立才, 颜红博. 基于PCA-SVM多生理信息融合的情绪识别方法[J]. 山东大学学报(工学版), 2014, 44(6): 70-76.
[14] 周咏梅1，杨佳能2,阳爱民2. 面向文本情感分析的中文情感词典构建方法[J]. 山东大学学报(工学版), 2013, 43(6): 27-33.
[15] 王昊,华继学,范晓诗. 基于双联支持向量机的入侵检测技术[J]. 山东大学学报(工学版), 2013, 43(6): 53-56.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2018 《山东大学学报(工学版)》编辑部　
地址: 济南市山大南路27号（250100）　 电话: 0531-88366735　 E-mail: xbgxb@sdu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发