基于灰狼群智能最优化的神经网络PM2.5浓度预测

doi:10.11772/j.issn.1001-9081.2017.10.2854

计算机应用 ›› 2017, Vol. 37 ›› Issue (10): 2854-2860.DOI: 10.11772/j.issn.1001-9081.2017.10.2854

基于灰狼群智能最优化的神经网络PM_2.5浓度预测

石峰¹, 楼文高^1,2, 张博³

1. 上海理工大学光电信息与计算机工程学院, 上海 200093;
2. 上海商学院信息与计算机学院, 上海 200235;
3. 上海理工大学出版印刷与艺术设计学院, 上海 200093

收稿日期:2017-04-20 修回日期:2017-06-22 出版日期:2017-10-10 发布日期:2017-10-16
通讯作者: 张博(1979-),男,天津人,副教授,博士,CCF会员,主要研究方向:网络新媒体、数字出版,E-mail:79379816@qq.com
作者简介:石峰(1992-),男,河南南阳人,硕士研究生,主要研究方向:人工神经网络、数据挖掘;楼文高(1964-),男,浙江杭州人,教授,博士,主要研究方向:金融工程、商务经济学、人工神经网络、数据挖掘;张博(1979-),男,天津人,副教授,博士,CCF会员,主要研究方向:网络新媒体、数字出版.
基金资助:
上海高校知识服务平台"上海市商贸服务业知识项目服务中心"建设项目（ZF1226）。

Neural network model for PM_2.5 concentration prediction by grey wolf optimizer algorithm

SHI Feng¹, LOU Wengao^1,2, ZHANG Bo³

1. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
2. Information and Computer Faculty, Shanghai Business School, Shanghai 200235, China;
3. College of Communication and Art Design, University of Shanghai for Science and Technology, Shanghai 200093, China

Received:2017-04-20 Revised:2017-06-22 Online:2017-10-10 Published:2017-10-16
Supported by:
This work is partially supported by the Construction Program of Shanghai Trade Service Knowledge Service Center in Shanghai University Knowledge Service Platform (ZF1226).

摘要/Abstract

摘要： 针对目前PM_2.5浓度测量成本高和测量过程繁杂等问题，建立了基于灰狼群智能最优化算法的神经网络预测模型。从非机理模型的角度，结合气象因素和空气污染物对上海市的PM_2.5浓度进行预测，并使用平均影响值分析了影响PM_2.5浓度的重要因素。使用灰狼群智能算法优化神经网络的过程中，引入了检验样本实时监控训练过程以避免发生"过训练"现象，确保建立的神经网络模型具有较好的泛化能力。实验结果表明：PM₁₀对PM_2.5的影响最为显著，其次是CO和前一天PM_2.5。选取2016年11月1日-12日的数据进行验证，其平均相对误差为13.46%，平均绝对误差为8μg/m³，与粒子群算法优化的神经网络、BP神经网络模型及支持向量回归（SVR）模型的误差相比，平均相对误差分别下降了约3个百分点、5个百分点和1个百分点。因此，使用灰狼算法优化的神经网络更适合上海市PM_2.5浓度的预测和空气质量的预报。

关键词: 灰狼优化算法, BP神经网络, PM_2.5浓度预测, 预测模型, 空气污染物

Abstract: Focusing on high cost and complicated process of the fine particulate matter (PM_2.5) measurement system, a neural network model based on grey wolf optimizer algorithm was established. From the perspective of non-mechanism model, the daily PM_2.5 concentration in Shanghai was forecasted with meteorological factors and air pollutants, and the important factors were analyzed by mean impact value. To avoid the "over training" and ensure the generalization ability, the validation datasets were used to monitor the training process. The experimental results show that the most significant factors that affecting the PM_2.5 concentration are PM₁₀, and then are the CO and the previous day's PM_2.5. Based on the datasets obtained from November 1, 2016 to November 12, the relative average error of the proposed model is 13.46%, the absolute average error is 8μg/m³; the relative average error of it is decreased by about 3 percentage points, 5 percentage points and 1 percentage points compared with the prediction models based on Particle Swarm Optimization (PSO), BP neural network and Support Vector Regression (SVR). The neural network model based on the grey wolf optimizer algorithm is more suitable for forecasting PM_2.5concentration and air quality in Shanghai.

Key words: grey wolf optimizer algorithm, BP neural network, PM_2.5 concentration prediction, prediction model, air pollutant

中图分类号:

石峰, 楼文高, 张博. 基于灰狼群智能最优化的神经网络PM_2.5浓度预测[J]. 计算机应用, 2017, 37(10): 2854-2860.

SHI Feng, LOU Wengao, ZHANG Bo. Neural network model for PM_2.5 concentration prediction by grey wolf optimizer algorithm[J]. Journal of Computer Applications, 2017, 37(10): 2854-2860.

参考文献

[1] 张兵, 杨俊, 宋昆仑, 等.大气污染的危害及其对经济社会的影响研究——以石家庄市为例[J]. 环境科学与管理, 2015, 40(7): 61-64. (ZAHNG B, YANG J, SONG K L, et al. Harm of air pollution and its impact on economy and society-taking Shijiazhuang city as an example[J]. Environmental Science and Management, 2015, 40(7): 61-64.)
[2] 马晓燕, 张志红.PM_2.5与哮喘关系的研究进展[J]. 环境与职业医学, 2015, 32(3): 279-283. (MA X Y, ZHANG Z H. Research progress on association between PM_2.5 and asthma[J]. Journal of Environmental and Occupational Medicine, 2015, 32(3): 279-283.)
[3] WANG Y, ZHUANG G, ZHANG X, et al. The ion chemistry, seasonal cycle, and sources of PM_2.5, and TSP aerosol in Shanghai[J]. Atmospheric Environment, 2006, 40(16): 2935-2952.
[4] 叶文波.宁波市大气可吸入颗粒物PM₁₀和PM_2.5的源解析研究[J]. 环境污染与防治, 2011, 33(9): 66-69. (YE W B. Study on source apportionment of PM₁₀ and PM_2.5 in ambient air of Ningbo[J]. Environmental Pollution and Control, 2011, 33(9): 66-69.)
[5] 贺祥, 林振山. 基于GAM模型分析影响因素交互作用对PM_2.5浓度变化的影响[J]. 环境科学, 2017,38(1): 22-32. (HE X, LIN Z S. Interactive effects of the influencing factors on the changes of PM_2.5 concentration based on gam model[J]. Environmental Science, 2017,38(1): 22-32.)
[6] 汤羹, 马宪国.上海气象因素对PM_2.5等大气污染物浓度的影响[J]. 能源研究与信息, 2016, 32(2): 71-74. (TANG G, MA X G. Influence of meteorological conditions in Shanghai on the concentration of PM_2.5and other air pollutants[J]. Energy Research and Information, 2016, 32(2): 71-74.)
[7] 张浩月, 王雪松, 陆克定, 等. 珠江三角洲秋季典型气象条件对O₃和PM₁₀污染的影响[J]. 北京大学学报(自然科学版), 2014, 50(3): 565-576. (ZHANG H Y, WANG X S, LU K D, et al. Impact of typical meteorological conditions on the O₃and PM₁₀ pollution episodes in the Pearl River Delta in autumn[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2014, 50(3): 565-576.)
[8] 王晓彦, 赵熠琳, 霍晓芹, 等. 基于数值模式的环境空气质量预报影响因素和改进方法[J]. 中国环境监测, 2016, 32(5): 1-7. (WANG X Y, ZHAO Y L, HUO X Q, et al. Discussion on the influence factors and improvement methods of ambient air quality forecasting based on numerical models[J]. Environmental Monitoring in China, 2016, 32(5): 1-7.)
[9] 王茜, 吴剑斌, 林燕芬. CMAQ模式及其修正技术在上海市PM_2.5预报中的应用检验[J]. 环境科学学报, 2015, 35(6): 1651-1656. (WANG Q, WU J B, LIN Y F. Implementation of a dynamic linear regression method on the CMAQ forecast of PM_2.5 in Shanghai[J]. Acta Scientiae Circumstantiae, 2015, 35(6): 1651-1656.)
[10] SYRAKOV D, PRODANOVA M, GEORGIEVA E, et al. Simulation of European air quality by WRF-CMAQ models using AQMEⅡ-2 infrastructure[J]. Journal of Computational & Applied Mathematics, 2016, 293 (C): 232-245.
[11] ZHANG Y, ZHANG X, WANG L, et al. Application of WRF/Chem over East Asia: part I, model evaluation and intercomparison with MM5/CMAQ[J]. Atmospheric Environment, 2016, 124: 285-300.
[12] 黄思, 唐晓, 徐文帅, 等. 利用多模式集合和多元线性回归改进北京PM₁₀预报[J]. 环境科学学报, 2015, 35(1): 56-64. (HUANG S, TANG X, XU W S, et al. Application of ensemble forecast and linear regression method in improving PM₁₀ forecast over Beijing areas[J]. Acta Scientiae Circumstantiae, 2015, 35(1): 56-64.)
[13] 王占山, 李晓倩, 王宗爽, 等. 空气质量模型CMAQ的国内外研究现状[J]. 环境科学与技术, 2013, 36(6): 386-391. (WANG Z S, LI X Q, WANG Z S, et al. Application status of Models-3/CMAQ in environmental management[J]. Environmental Science & Technology, 2013, 36(6): 386-391.)
[14] 余辉, 袁晶, 于旭耀, 等. 基于ARMAX的PM_2.5小时浓度跟踪预测模型[J]. 天津大学学报 (自然科学与工程技术版), 2017, 50(1): 105-111. (YU H, YUAN J, YU X Y, et al. Tracking prediction model for PM_2.5 hourly concentration based on ARMAX[J]. Journal of Tianjin University (Science and Technology), 2017, 50(1): 105-111.)
[15] 彭斯俊, 沈加超, 朱雪.基于ARIMA模型的PM_2.5预测[J]. 安全与环境工程, 2014, 21(6): 125-128. (PENG S J, SHEN J C, ZHU X. Forecast of PM_2.5 based on the arima model[J]. Safety and Environmental Engineering, 2014, 21(6): 125-128.)
[16] 张玉丽, 何玉, 朱家明.基于多元线性回归模型PM_2.5预测问题的研究[J]. 安徽科技学院学报, 2016, 30(3): 92-97. (ZAHNG Y L, HE Y, ZHU J M. Study of the prediction of PM_2.5 based on the multivariate linear regression model[J]. Journal of Anhui Science and Technology University, 2016, 30(3): 92-97.)
[17] TAI A P K, MICKLEY L J, JACOB D J. Correlations between fine particulate matter (PM_2.5) and meteorological variables in the United States: implications for the sensitivity of PM_2.5 to climate change[J]. Atmospheric Environment, 2010, 44(32): 3976-3984.
[18] 艾洪福, 潘贺, 李迎斌. 关于空气雾霾PM_2.5含量预测优化研究[J]. 计算机仿真, 2017, 34(1): 392-395. (AI H F, PAN H, LI Y B. Research on optimization of PM_2.5 content prediction in air haze[J]. Computer Simulation, 2017, 34(1): 392-395.)
[19] 戴华炜, 陈小旋, 孙晓通. 基于GM (1, 1) 模型群的深圳市大气污染物灰色预测[J]. 数学的实践与认识, 2014, 44(1): 131-136. (DAI H W, CHEN X X, SUN X T. The gray prediction model of the atmospheric pollutants in Shenzhen based on GM (1, 1) [J]. Mathematics in Practice and Theory, 2014, 44(1): 131-136.)
[20] 杨忠, 童楚东, 俞杰, 等. 加权因子的PSO-SVR区域空气PM_2.5浓度预报方法[J]. 计算机应用研究, 2017(2): 405-408. (YANG Z, TONG C D, YU J, et al. Regional PM_2.5 concentration prediction method of PSO-SVR model with weighting factors[J]. Application Research of Computers, 2017(2): 405-408.)
[21] 李龙, 马磊, 贺建峰, 等. 基于特征向量的最小二乘支持向量机PM_2.5浓度预测模型[J]. 计算机应用, 2014, 34(8): 2212-2216. (LI L, MA L, HE J F, et al. PM_2.5concentration prediction model of least squares support vector machine based on feature vector[J]. Journal of Computer Applications, 2014, 34(8): 2212-2216.)
[22] 葛考, 时贝贝, 徐康耀, 等.基于BAPSO-PNN神经网络算法的空气质量评价研究[J]. 环境工程, 2015, 33: 676-681. (GE K, SHI B B, XU K Y, et al. Air quality assement based on BAPSO-PNN neural network algorithm [J]. Environmental Engineering, 2015, 33: 676-681.)
[23] 杨云, 付彦丽.基于T-S模型模糊神经网络的PM_2.5质量浓度预测[J]. 陕西科技大学学报, 2015, 33(6): 162-166. (YANG Y, FU Y L. The prediction of mass concentration of PM_2.5 based on T-S fuzzy neural network[J]. Journal of Shaanxi University of Science and Technology, 2015, 33(6): 162-166.)
[24] 马天成, 刘大铭, 李雪洁, 等.基于改进型PSO的模糊神经网络PM_2.5浓度预测[J]. 计算机工程与设计, 2014, 35(9): 3258-3262. (MA T C, LIU D M, LI X J, et al. Improved particle swarm optimization based fuzzy neural network for PM_2.5 concentration prediction[J]. Computer Engineering and Design, 2014, 35(9): 3258-3262.)
[25] 荆涛, 李霖, 于文柱, 等.t分布受控遗传算法优化BP神经网络的PM_2.5质量浓度预测[J]. 中国环境监测, 2015, 31(4): 100-105. (JIN T, LI L, YU W Z, et al. Prediction of PM_2.5 mass concentration based on BP neural network optimized by t-distribution controlled genetic algorithm[J]. Environmental Monitoring in China, 2015, 31(4): 100-105.)
[26] 谢超, 马民涛, 于肖肖.多种神经网络在华北西部区域城市空气质量预测中的应用[J]. 环境工程学报, 2015, 9(12): 6005-6009. (XIE C, MA M T, YU X X. Forecasting model of air pollution index based on multi-artificial neural network in western region of Northern China[J]. Chinese Journal of Environmental Engineering, 2015, 9(12): 6005-6009.)
[27] 李嵩, 王冀, 张丹闯, 等. 大气PM_2.5污染指数预测优化模型仿真分析[J]. 计算机仿真, 2015, 32(12): 400-403. (LI S, WANG J, ZHANG D C, et al. Simulation analysis of prediction optimization model for atmospheric PM_2.5 pollution index[J]. Computer Simulation, 2015, 32(12): 400-403.)
[28] MISHRA D, GOYAL P, UPADHYAY A. Artificial intelligence based approach to forecast PM_2.5, during haze episodes: a case study of Delhi, India[J]. Atmospheric Environment, 2015, 102: 239-248.
[29] HAYKIN S. Neural Networks: a Comprehensive Foundation [M]. 2rd ed. Upper Saddle River, NJ: Prentice Hall Press, 1998.
[30] BISHOP C M. Neural Networks for Pattern Recognition[M]. Oxford: Oxford University Press, 1995.
[31] ZHANG G, PATUWO B E, HU M Y. Forecasting with artificial neural networks: the state of the art[J]. International Journal of Forecasting, 1998, 14(1): 35-62.
[32] 楼文高, 王延政.基于BP网络的水质综合评价模型及其应用[J]. 环境工程学报, 2003, 4(8): 23-26. (LOU W G, WANG Y Z. Water quality comprehensive assessment model using BP networks and its applications[J]. Chinese Journal of Environmental Engineering, 2003, 4(8): 23-26.)
[33] 楼文高.实现BP神经网络从理论到实践的跨越[J]. 哈尔滨工程大学学报, 2006, 27: 59-63. (LOU W G. Realization of BP neural network from theory to practice[J]. Journal of Harbin Engineering University, 2006, 27: 59-63.)
[34] MIRJALILI S, MIRJALILI S M, LEWIS A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69(3): 46-61.
[35] MIRJALILI S. How effective is the grey wolf optimizer in training multi-layer perceptrons[J]. Applied Intelligence, 2015, 43(1): 150-161.
[36] 葛跃, 王明新, 孙向武, 等. 基于增强回归树的城市PM_2.5日均值变化分析: 以常州为例[J]. 环境科学, 2017, 38(2): 485-494. (GE Y, WANG M X, SUN X W, et al. Variation analysis of daily PM_2.5 concentrations based on boosted regression tree: a case study in Changzhou[J]. Environmental Science, 2017, 38(2): 485-494.)
[37] HECHT-NIELSEN R. Theory of the backpropagation neural network [M]// WECHSLER H. Neural Networks for Perception. New York: Academic Press, 1991: 65-93.
[38] 楼文高, 乔龙. 基于神经网络的金融风险预警模型及其实证研究[J]. 金融论坛, 2011(11): 52-61. (LOU W G, QIAO L. Early warning model of financial risks and empirical study based on neural network[J]. Finance Forum, 2011(11): 52-61.)
[39] DOMBI G W, NANDI P, SAXE J M, et al. Prediction of rib fracture injury outcome by an artificial neural network [J]. Journal of Trauma, 1995, 39(5): 915-921.

基于灰狼群智能最优化的神经网络PM_2.5浓度预测

Neural network model for PM_2.5 concentration prediction by grey wolf optimizer algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	徐国保, 陈媛晓, 王骥. 基于图卷积网络的药物靶标关联预测算法[J]. 计算机应用, 2021, 41(5): 1522-1526.
[2]	许飞雪, 刘勤明, 欧阳海玲, 叶春明. 基于性能合同的多部件系统维修策略模型优化[J]. 计算机应用, 2021, 41(4): 1184-1191.
[3]	郝芃斐, 池瑞, 屈志坚, 涂宏斌, 池学鑫, 张地友. 求解铁路物流配送中心选址问题的改进灰狼优化算法[J]. 计算机应用, 2021, 41(10): 2905-2911.
[4]	刘娟, 黄细霞, 刘晓丽. 基于栈式自编码网络的风机叶片结冰预测[J]. 计算机应用, 2019, 39(5): 1547-1550.
[5]	郑卓然, 郑向伟, 田杰. 基于惩罚误差矩阵的同步预测无线体域网节能方法[J]. 计算机应用, 2019, 39(2): 513-517.
[6]	于家斌, 尚方方, 王小艺, 许继平, 王立, 张慧妍, 郑蕾. 基于遗传算法改进的一阶滞后滤波和长短期记忆网络的蓝藻水华预测方法[J]. 计算机应用, 2018, 38(7): 2119-2123.
[7]	商建东, 李盼乐, 刘润杰, 李润川. 基于加权时变泊松模型的出租车载客点推荐模型[J]. 计算机应用, 2018, 38(4): 923-927.
[8]	何春, 李琦, 吴让好, 刘邦欣. 基于故障传播的模块化BP神经网络电路故障诊断[J]. 计算机应用, 2018, 38(2): 602-609.
[9]	陈万志, 李东哲. 结合白名单过滤和神经网络的工业控制网络入侵检测方法[J]. 计算机应用, 2018, 38(2): 363-369.
[10]	陈煜, 蒋伟, 周继恩. 基于预测模型的轨迹数据压缩方法[J]. 计算机应用, 2018, 38(1): 171-175.
[11]	王军, 费凯, 程勇. 基于改进的Adaboost-BP模型在降水中的预测[J]. 计算机应用, 2017, 37(9): 2689-2693.
[12]	李宁, 李刚, 邓中亮. 改进灰狼算法在土壤墒情监测预测系统中的应用[J]. 计算机应用, 2017, 37(4): 1202-1206.
[13]	陈闯, Ryad Chellali, 邢尹. 采用动态权重和概率扰动策略改进的灰狼优化算法[J]. 计算机应用, 2017, 37(12): 3493-3497.
[14]	徐达宇, LIU Renping. 基于改进自组织临界优化的元启发式灰狼优化算法[J]. 计算机应用, 2016, 36(6): 1588-1593.
[15]	刘军, 白雪. 基于梯度方向直方图与高斯金字塔的车牌模糊汉字识别方法[J]. 计算机应用, 2016, 36(2): 586-590.

基于灰狼群智能最优化的神经网络PM2.5浓度预测

Neural network model for PM2.5 concentration prediction by grey wolf optimizer algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

基于灰狼群智能最优化的神经网络PM_2.5浓度预测

Neural network model for PM_2.5 concentration prediction by grey wolf optimizer algorithm