第4期 周杉杉，等：基于自组织递归模糊神经网络的PM2.5浓度预测 ·515· 统辨识实验证明了该预测模型的有效性，最后将 vironment,2015,102:239-248 该模型用于实际问题PM2.5浓度的预测。实验结 [11]QIAO Junfei,CAI Jie,HAN Honggui,et al.Predicting 果表明该模型不仅能够获得较为精简的网络结 PM2s concentrations at a regional background station us- 构，而且预测精度有所提高，为PM2.5浓度预测 ing second order self-organizing fuzzy neural network[J]. 提供了一种有效的方法。 Atmosphere,2017,8(1):10. [12]HAN Honggui,LI Ying,GUO Yanan,et al.A soft com- 参考文献： puting method to predict sludge volume index based on a recurrent self-organizing neural network[J].Applied soft [1]TAO Minghui,CHEN Liangfu,WANG Zifeng,et al.A computing,2016.38:477-486. study of urban pollution and haze clouds over northern [13]HAN Honggui,WANG Lidan,QIAO Junfei,et al.A China during the dusty season based on satellite and sur- face observations[J].Atmospheric environment,2014,82: spiking-based mechanism for self-organizing RBF neural networks[C]//2014 International Joint Conference on 183-192 [2]QIAO Liping.CAI Jing,WANG Hongqi,et al.PM2.5 con- Neural Networks (IJCNN).Beijing,China,2014: 3775-3782 stituents and hospital emergency-room visits in Shanghai, [14]HAN Honggui,WU Xiaolong,QIAO Junfei.Nonlinear China[J].Environmental science and technology,2014, 48(17):10406-10414. systems modeling based on self-organizing fuzzy-neural- [3]XIAO S,WANG Q Y,CAO JJ,et al.Long-term trends in network with adaptive computation algorithm[J].IEEE visibility and impacts of aerosol composition on visibility transactions on cybernetics,2014,44(4):554-564. impairment in Baoji,China[J].Atmospheric research, [15]LENG Gang,MCGINNITY T M,PRASAD G.An ap- 2014,149:88-95. proach for on-line extraction of fuzzy rules using a self- [4]SAIDE P E.CARMICHAEL G R,SPAK S N,et al.Fore- organising fuzzy neural network[J].Fuzzy sets and sys- casting urban PM10 and PM2.5 pollution episodes in very tems,2005,150(2):211-243 stable nocturnal conditions and complex terrain using [16]WU Shiqian,ER M J.Dynamic fuzzy neural networks-a WRF-Chem CO tracer model[J].Atmospheric environ- novel approach to function approximation[J].IEEE trans- ment.2011.45(16):2769-2780 actions on systems,man,and cybernetics,part B(cyber- [5]RICCIO A,CHIANESE E,AGRILLO G,et al.Source ap- netics).2000.30(2):358-364. portion of atmospheric particulate matter:a joint [17]WU Shiqian,ER M J,GAO Yang.A fast approach for Eulerian/Lagrangian approach[J].Environmental science automatic generation of fuzzy rules by generalized dy- and pollution research,2014,21(23):13160-13168. namic fuzzy neural networks[J].IEEE transactions on [6]CHEN Yuanyuan,SHI Runhe,SHU Shijie,et al.En- fuzzy systems,.2001,94):578-594. semble and enhanced PM10 concentration forecast model [18]陈冠益，张雯，侯立安，等.天津蓟县夏季PM2.5污染特 based on stepwise regression and wavelet analysis[J].At- 征及影响因素).天津大学学报：自然科学与工程技术 mospheric environment,2013,74:346-359. 版，2015,48(2)：95-102 [7]ELBAYOUMI M,RAMLI N A,MD YUSOF N FF,et al. CHEN Guanyi,ZHANG Wen,HOU Lian,et al.Pollution Multivariate methods for indoor PM10 and PM2.5 model- characteristics and influence factors of PM2.5 in summer ling in naturally ventilated schools buildings[J].Atmo- in Jixian county of Tianjin[J].Journal of Tianjin uni- spheric environment,2014,94:11-21. versity:science and technology,2015,48(2):95-102. [8]ORDIERES J B.VERGARA E P,CAPUZ R S,et al. [19]ZHENG Yu,YI Xiuwen,LI Ming,et al.Forecasting fine- Neural network prediction model for fine particulate mat- grained air quality based on big data[C]//Proceedings of ter (PM2.5)on the US-Mexico border in El Paso (Texas) the 21th ACM SIGKDD International Conference on and Ciudad Juarez(Chihuahua)[J].Environmental model- Knowledge Discovery and Data Mining.New York,NY, ling and software,2005,20(5):547-559. USA.2015:2267-2276 [9]XU Zhao,XIA Xiaopeng,LIU Xiangnan,et al.Combin- [20]AZID A,JUAHIR H,TORIMAN M E,et al.Prediction ing DMSP/OLS nighttime light with echo state network for of the level of air pollution using principal component prediction of daily PM2.5 average concentrations in analysis and artificial neural network techniques:A case Shanghai,China[J].Atmosphere,2015,6(10):1507-1520. study in Malaysia[J].Water,air,and soil pollution,2014, [10]MISHRA D.GOYAL P.UPADHYAY A.Artificial intel- 225:2063. ligence based approach to forecast PM2.5 during haze [21]VOUKANTSIS D,KARATZAS K,KUKKONEN J,et al. episodes:a case study of Delhi,India[J].Atmospheric en- Intercomparison of air quality data using principal com-统辨识实验证明了该预测模型的有效性，最后将 该模型用于实际问题 PM2.5 浓度的预测。实验结 果表明该模型不仅能够获得较为精简的网络结 构，而且预测精度有所提高，为 PM2.5 浓度预测 提供了一种有效的方法。 参考文献： TAO Minghui, CHEN Liangfu, WANG Zifeng, et al. A study of urban pollution and haze clouds over northern China during the dusty season based on satellite and sur￾face observations[J]. Atmospheric environment, 2014, 82: 183–192. [1] QIAO Liping, CAI Jing, WANG Hongqi, et al. PM2.5 con￾stituents and hospital emergency-room visits in Shanghai, China[J]. Environmental science and technology, 2014, 48(17): 10406–10414. [2] XIAO S, WANG Q Y, CAO J J, et al. Long-term trends in visibility and impacts of aerosol composition on visibility impairment in Baoji, China[J]. Atmospheric research, 2014, 149: 88–95. [3] SAIDE P E, CARMICHAEL G R, SPAK S N, et al. Fore￾casting urban PM10 and PM2.5 pollution episodes in very stable nocturnal conditions and complex terrain using WRF–Chem CO tracer model[J]. Atmospheric environ￾ment, 2011, 45(16): 2769–2780. [4] RICCIO A, CHIANESE E, AGRILLO G, et al. Source ap￾portion of atmospheric particulate matter: a joint Eulerian/Lagrangian approach[J]. Environmental science and pollution research, 2014, 21(23): 13160–13168. [5] CHEN Yuanyuan, SHI Runhe, SHU Shijie, et al. En￾semble and enhanced PM10 concentration forecast model based on stepwise regression and wavelet analysis[J]. At￾mospheric environment, 2013, 74: 346–359. [6] ELBAYOUMI M, RAMLI N A, MD YUSOF N F F, et al. Multivariate methods for indoor PM10 and PM2.5 model￾ling in naturally ventilated schools buildings[J]. Atmo￾spheric environment, 2014, 94: 11–21. [7] ORDIERES J B, VERGARA E P, CAPUZ R S, et al. Neural network prediction model for fine particulate mat￾ter (PM2.5) on the US–Mexico border in El Paso (Texas) and Ciudad Juárez (Chihuahua)[J]. Environmental model￾ling and software, 2005, 20(5): 547–559. [8] XU Zhao, XIA Xiaopeng, LIU Xiangnan, et al. Combin￾ing DMSP/OLS nighttime light with echo state network for prediction of daily PM2.5 average concentrations in Shanghai, China[J]. Atmosphere, 2015, 6(10): 1507–1520. [9] MISHRA D, GOYAL P, UPADHYAY A. Artificial intel￾ligence based approach to forecast PM2.5 during haze episodes: a case study of Delhi, India[J]. Atmospheric en- [10] vironment, 2015, 102: 239–248. QIAO Junfei, CAI Jie, HAN Honggui, et al. Predicting PM2.5 concentrations at a regional background station us￾ing second order self-organizing fuzzy neural network[J]. Atmosphere, 2017, 8(1): 10. [11] HAN Honggui, LI Ying, GUO Yanan, et al. A soft com￾puting method to predict sludge volume index based on a recurrent self-organizing neural network[J]. Applied soft computing, 2016, 38: 477–486. [12] HAN Honggui, WANG Lidan, QIAO Junfei, et al. A spiking-based mechanism for self-organizing RBF neural networks[C]//2014 International Joint Conference on Neural Networks (IJCNN). Beijing, China, 2014: 3775–3782. [13] HAN Honggui, WU Xiaolong, QIAO Junfei. Nonlinear systems modeling based on self-organizing fuzzy-neural￾network with adaptive computation algorithm[J]. IEEE transactions on cybernetics, 2014, 44(4): 554–564. [14] LENG Gang, MCGINNITY T M, PRASAD G. An ap￾proach for on-line extraction of fuzzy rules using a self￾organising fuzzy neural network[J]. Fuzzy sets and sys￾tems, 2005, 150(2): 211–243. [15] WU Shiqian, ER M J. Dynamic fuzzy neural networks-a novel approach to function approximation[J]. IEEE trans￾actions on systems, man, and cybernetics, part B (cyber￾netics), 2000, 30(2): 358–364. [16] WU Shiqian, ER M J, GAO Yang. A fast approach for automatic generation of fuzzy rules by generalized dy￾namic fuzzy neural networks[J]. IEEE transactions on fuzzy systems, 2001, 9(4): 578–594. [17] 陈冠益, 张雯, 侯立安, 等. 天津蓟县夏季 PM2.5 污染特 征及影响因素 [J]. 天津大学学报: 自然科学与工程技术 版, 2015, 48(2): 95–102. CHEN Guanyi, ZHANG Wen, HOU Lian, et al. Pollution characteristics and influence factors of PM2.5 in summer in Jixian county of Tianjin[J]. Journal of Tianjin uni￾versity: science and technology, 2015, 48(2): 95–102. [18] ZHENG Yu, YI Xiuwen, LI Ming, et al. Forecasting fine￾grained air quality based on big data[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York, NY, USA, 2015: 2267–2276. [19] AZID A, JUAHIR H, TORIMAN M E, et al. Prediction of the level of air pollution using principal component analysis and artificial neural network techniques: A case study in Malaysia[J]. Water, air, and soil pollution, 2014, 225: 2063. [20] VOUKANTSIS D, KARATZAS K, KUKKONEN J, et al. Intercomparison of air quality data using principal com- [21] 第 4 期 周杉杉，等：基于自组织递归模糊神经网络的 PM2.5 浓度预测 ·515·