人工智能技术在大气环境领域的应用

梁常德; 尹民; 胡清; 游勇; 黄芊蕙; 许盛彬

doi:10.13205/j.hjgc.202504010

人工智能技术在大气环境领域的应用

doi: 10.13205/j.hjgc.202504010

梁常德¹,
尹民¹,
胡清²,
游勇¹,
黄芊蕙²,
许盛彬^2, ,

1. 深圳市生态环境智能管控中心,广东深圳 518000;
2. 南方科技大学环境科学与工程学院,广东深圳 518000

基金项目:

国家重点研发计划“区块链边缘计算原型系统设计和应用开发”（2023YFB2704604）；深圳市国家智能社会治理实验特色基地（环境治理）项目（SZDL2023001177）

详细信息

作者简介:
梁常德（1980-），男，硕士，高级工程师，主要研究方向为生态环境大数据技术与应用。836780343@qq.com

通讯作者:
许盛彬（1990-），男，硕士，中级工程师，主要研究方向为生态环境大数据技术与应用。xusb@mail.sustech.edu.cn

计量
- 文章访问数: 135
- HTML全文浏览量: 20
- PDF下载量: 2
- 被引次数: 0
出版历程
- 收稿日期: 2024-06-27
- 录用日期: 2024-10-22
- 修回日期: 2024-08-22
- 刊出日期: 2025-04-01

Applications of artificial intelligence technology in the atmospheric environment field

1. Shenzhen Ecological Environment Intelligent Management and Control Center, Shenzhen 518000, China;
2. School of Environmental Science and Engineering, Southern University of Science and Technology, Shenzhen 518000, China

摘要

摘要: 近年来，人工智能技术迅速发展，引起了各领域研究人员的关注，在农业、气候、安全、环境等不同学科领域均取得了丰硕成果。首先对模糊逻辑、遗传算法、人工神经网络、支持向量机、轻量级梯度提升机等人工智能技术的基本原理、技术特征和适用范围进行了总结，重点阐述混合智能系统的技术先进性和优势；其次，以大气环境领域为例，调研了人工智能技术在环境领域的应用进展，借助具体案例说明了常见的输入参数类型和性能评价指标；最后，从环境要素数据集、模型训练方式、算法自有缺陷3个方面分析了现阶段人工智能技术在大气环境领域存在的挑战，提出了人工智能技术在大气环境领域未来的应用发展趋势主要为混合智能系统组合研究、多环境要素综合分析及其与环保平台的协同应用。
- 人工智能 /
- 大气环境领域 /
- 模糊逻辑 /
- 遗传算法 /
- 人工神经网络
Abstract: In recent years， the rapid development of artificial intelligence technology has attracted the attention of researchers in most fields， and fruitful achievements have been made in agriculture， climate， security， and the environment. Firstly， the review summarized the basic principles， technical characteristics， and application scope of artificial intelligence technologies such as fuzzy logic， genetic algorithms， artificial neural networks， XGBoost， LightGBM， etc. Moreover， it focused on the technical advancements and advantages of hybrid intelligent systems which combined artificial neural networks and different artificial intelligence technologies. Secondly， taking the field of atmospheric environment as an example， the application progress of artificial intelligence technology in environmental fields was investigated. Especially， by summarizing the common input parameter types and performance evaluation indexes for the prediction of different air pollutants， the prediction applications in the field of atmospheric environment of single artificial intelligence models and hybrid artificial intelligence models were illustrated with specific cases. Generally， artificial model prediction accuracy is affected by various factors， including the input parameters and model types. Compared to single neural network models， hybrid intelligent models have a relatively higher predictive performance. Finally， the challenges of artificial intelligence in the atmospheric environment field were analyzed from environmental element data sets， model training methods， and algorithm defects. In the future， the applications of artificial intelligence technology in the atmospheric environment field should be improved from the following aspects： hybrid intelligent system combinations， comprehensive analysis of complicated environmental elements， and cooperative working with environmental protection platforms.
- artificial intelligence /
- atmospheric environmental field /
- fuzzy logic /
- genetic algorithm /
- artificial neural network

HTML全文

1 大气环境领域的人工智能分类树

注： GA-ANN为遗传算法与人工神经网络；ES-ANN为专家系统与人工神经网络；ANN-ANN为由2种或以上ANN技术组合的混合人工神经网络系统；PSO-ANN为粒子群优化神经网络； GRU为循环神经网络。

1. A classification tree of AI for atmospheric environmental fields

下载: 全尺寸图片幻灯片

3 人工智能算法在空气污染预测中的建模流程^［29］

3. Modeling process flowchart for AI in air pollution forecasting^［29］

下载: 全尺寸图片幻灯片

1 不同人工智能技术简介

1. Introduction to different AI technologies

类型	机制	优势	短板
MLP	各层神经元全连接，无层内连接，无跨层连接	非线性映射、有利于并行运算	泛化能力差、处理多维数据能力差
RBFN	只有1个隐藏层，具有局部映射的特征	具有极快的学习收敛速度，具有最佳逼近性能和全局最优特性	解释性差，数据不充分时无法工作，难以确定隐藏层节点数、节点中心和宽度
RNN	隐藏层节点输出取决于当前节点输入和上个节点值，实现层内神经元连接和循环	提取时序特征能力强、泛化能力相对较好，适合解决时间依赖性问题^[24]	输入输出序列不同、处理长期依赖精度下降，存在梯度爆炸和梯度消失问题^[25]
LSTM	RNN的改良体，细胞状态，只有少量线性交互	在序列建模问题上有一定的优势，具有长时记忆功能	并行处理存在劣势，训练时间较长，效率较低，处理小规模数据效果不理想
CNN	神经元之间局部连接，卷积核特征提取^[26]	稀疏连接、权值共享、计算量小、无需复杂的预处理^[26]	输入图像尺寸固定
SVM	基于结构化风险最小化原则，用二次元求解支持向量	高鲁棒性、全局搜索能力，所需数据量小，过拟合概率小^[27]	大数据集效率低^[28]，批量运行时内存及CPU需求量大
GA	模拟了生物进化论中的自然选择和遗传机制	具有良好的全局优化能力，与其他模型方法结合较灵活^[28]	计算量大^[28]，在达到理想效果前需要多次试错
FL	一种类似于人类推理的推理方法，涉及YES和NO之间的所有中间可能性	鲁棒性强，较强的容错能力，构建任意复杂性的非线性函数的能力^[29]	精度较低，难定义控制目标^[28]
XGBoost	基于预排序方法的决策树算法	支持并行计算，准确精度较高^[30]，可扩展性高^[31]，可进行缺失值处理	内存空间消耗大
LightGBM	弱分类器（决策树）迭代训练，利用网络通信算法来优化并行学习	逐叶树木生长式算法，训练时间短，支持高效并行处理，内存消耗少^[32]	可能发生过拟合

下载: 导出CSV

2 不同人工神经网络结构特征

2. Structural characteristics of different ANNs


a-MLP	b-RBFN	c-CNN

d-RNN	e-LSTM

下载: 导出CSV

2 不同混合人工智能系统简介

2. Introduction to different AI technologies

类型	机制	组合优势	文献
FNN	模糊逻辑理论、遗传算法、神经网络之间的协同协作	扩大处理范围（可处理定量和定性信息），提高信息处理速度	[65]
ES-ANN	融合ANN与ES系统	响应速度快，抗干扰、知识自动化获取能力及解释能力较强	[66，67]
ANFIS	基于ANN和FL的数据驱动建模技术，输入与输出映射通过基于网络形式的模糊规则连接	灵活性、处理速度和适应能力得到优化	[68-71]
PSO-ANN	包括内外两层，外层为PSO算法，内层为ANN网络，实现协同搜索计算	克服了局部最小值问题	[72]
GA-ANN	将GA与神经网络结合以迭代优化神经网络参数	网络配置得到优化，预测效率、灵活性及稳定性较高	[73]

下载: 导出CSV

3 人工智能技术在大气环境领域的应用

3. Applications of AI technology in the atmospheric environment field

年份	类型	研究对象	输入参数	输出参数	性能评价指标	文献
2009	GA	空气污染预测	PM₁₀样本数据	PM₁₀	RMSE	[89]
2012	MLP、RBFN、GRNN	城市空气质量	气温、相对湿度、风速、空气质量监测数据	可吸入颗粒物、SO₂、NO₂	RMSE、MAE	[90]
2016	ANFIS	空气质量预测	气压、温度、湿度、能见度、露点、风速、降水等	SO₂、NO₂、CO、O₃、PM₁₀等	R²、NMRS等	[91]
2016	MLP	空气中CO浓度	温度、风速、湿度、CO、水分	CO浓度	MAE、R²	[92]
2018	ARIMAX-MLP	PM_2.5预测	温度范围、湿度范围、日照时长、风速、风向、NO₂、SO₂、O₃、CO、PM_2.5等	PM_2.5	RMSE、MAE、MAPE	[93]
2019	RNN、RF	空气污染物预测	气象学变量（温度、温度露点差、相对湿度、降水、风速、风向、大气压力和城市热岛）、大气污染物	污染物浓度	R、RSEM	[94]
2020	CNN	预测马来西亚巴生市空气污染指数	PM₁₀、CO、O₃、NO₂、SO₂等	API	MAE、RMSE、MAPE、R²	[86]
2020	MLP	O₃预测	温度、湿度、风速、风向、O₃等	O₃浓度	R²、MAE、RMSE	[95]
2020	CNN-LSTM	PM_2.5短期预测	气温、风速、风向、气压、累计降雨量等气象因素	PM_2.5	RMSE、MAE、MAPE	[96]
2021	MLP、LSTM	空气质量评估和污染预测	PM₁₀、温度、相对湿度、风速、风向	未来某段时间空气质量	MAE、MSE、MAPE	[97]
2021	LSTM	大气不同污染物相关性分析	O₃、温度、NO_x、光合有效辐射等	异戊二烯和PM₁₀预测	—	[98]
2022	FL	城市空气质量决策支持	PM₁₀污染水平和大气稳定性信息	不同领域空气质量评价	—	[99]
2024	基于应用策略的LSTM（ASLSTM）	PM_2.5短期预测	PM_2.5、O₃、PM₁₀、CO、SO₂、NO、相对湿度、风速、风向、气温	污染物浓度	RMSE、MAE	[79]

下载: 导出CSV

参考文献(99)

[1]	LIN Z Q. A short biography of John McCarthy［J］. Computer Science，1993（4）:32-33 林作铨. 约翰·麦卡锡小传［J］. 计算机科学，1993（4）:32-3
[2]	CHAMBERS D M R C M，THORNBURG L G，et al. Distinguishing petroleum（crude oil and fuel）from smoke exposure within populations based on the relative blood levels of benzene，toluene，ethylbenzene，and xylenes（BTEX），styrene and 2，5-dimethylfuran by pattern recognition using artificial neural networks［J］. Environment Science Technology，2018，625（5）:75-88.
[3]	HONG H，TSANGARATOS P，ILIA I，et al. Application of fuzzy weight of evidence and data mining techniques in construction of flood susceptibility map of Poyang County，China［J］. Science of the Total Environment，2018，625:575-588.
[4]	LESNIK K L，LIU H. Predicting microbial fuel cell biofilm communities and bioreactor performance using artificial neural networks［J］. Environmental Science& Technology，2017，51（18）:10881-10892.
[5]	NABAVI P A，RAFIEE S，MOHTASEBI S S，et al. Integration of artificial intelligence methods and life cycle assessment to predict energy output and environmental impacts of paddy production［J］. Science of the Total Environment，2018，631:1279-1294.
[6]	ZHANG R，CHEN Z Y，XU L J，et al. Meteorological drought forecasting based on a statistical model with machine learning techniques in Shaanxi province，China［J］. Science of the Total Environment，2019，665:338-346.
[7]	MA S Y. An introduction to the application of artificial intelligence in environmental engineering［J］. Science and Informatization，2020. 马思远. 浅谈人工智能在环境工程中的应用［J］. 科学与信息化，2020.
[8]	PALANI S，TKALICH P，BALASUBRAMANIAN R，et al. ANN application for prediction of atmospheric nitrogen deposition to aquatic ecosystems［J］. Marine Pollution Bulletin，2011，62（6）:1198-1206.
[9]	CSÁBRÁGI A，MOLNÁR S，TANOS P，et al. Estimation of dissolved oxygen in riverine ecosystems:comparison of differently optimized neural networks［J］. Ecological Engineering，2019，138:298-309.
[10]	FAROOQ M U，ZAFAR A M，RAHEEM W，et al. Assessment of algorithm performance on predicting total dissolved solids using artificial neural network and multiple linear regression for the groundwater data［J］. Water，2022，14（13）:2002.
[11]	TAOUFIK N，BOUMYA W，ACHAK M，et al. Comparative analysis of response surface methodology and some artificial intelligence models in the prediction of methyl green degradation by Fenton process［J］. International Journal of Environmental Analytical Chemistry，2021:1-18.
[12]	BECK M B，LATTEN A，TONG R M. Modelling and operational control of the activated sludge process in wastewater treatment［J］. 1978.
[13]	ENBUTSU I，BABA K，HARA N，et al. Integration of multi AI paradigms for intelligent operation support systems-fuzzy rule extraction from a neural network［J］. Water Science and Technology，1993，28（11-12）:333-340.
[14]	FAN M，HU J，CAO R，et al. A review on experimental design for pollutants removal in water treatment with the aid of artificial intelligence［J］. Chemosphere，2018，200:330-343.
[15]	ZHAO L，DAI T，QIAO Z，et al. Application of artificial intelligence to wastewater treatment:a bibliometric analysis and systematic review of technology，economy，management，and wastewater reuse［J］. Process Safety and Environmental Protection，2020，133:169-182.
[16]	BOGER Z. Application of neural networks to water and wastewater treatment plant operation［J］. ISA transactions，1992，31（1）:25-33.
[17]	SCHLEITER I M，BORCHARDT D，WAGNER R，et al. Modelling water quality，bioindication and population dynamics in lotic ecosystems using neural networks［J］. Ecological Modelling，1999，120（2/3）:271-286.
[18]	TRAUTMANN T，DENŒUX T. A constructive algorithm for SOM applied to water quality monitoring［J］. Intelligent Engineering Systems through Artificial Neural Networks，1994，4:17-22.
[19]	WALLEY W，FONTAMA V. Neural network predictors of average score per taxon and number of families at unpolluted river sites in Great Britain［J］. Water Research，1998，32（3）:613-622.
[20]	MAIER H R，DANDY G C. The use of artificial neural networks for the prediction of water quality parameters［J］. Water Resources Research，1996，32（4）:1013-1022.
[21]	SOHN S H，OH S C，YEO Y K. Prediction of air pollutants by using an artificial neural network［J］. Korean Journal of Chemical Engineering，1999，16（3）:382-387.
[22]	YANG C C，PRASHER S O，MEHUYS G R. An artificial neural network to estimate soil temperature［J］. Canadian Journal of Soil Science，1997，77（3）:421-429.
[23]	ZHANG J P，WU S X. An application of artificial neural network to traffic noise prediction［J］. Acta Scientiae Circumstantiae，1998（5）:25-31. 张继萍，吴硕贤. 人工神经网络在道路交通噪声预测中的应用［J］. 环境科学学报，1998（5）:25-31.
[24]	HAYKIN S. Neural networks and learning machines，3/E［M］. Pearson Education India，2009.
[25]	Pascanu R，Mikolov T，Bengio Y. On the difficulty of training recurrent neural networks［C］// International Conference on Machine Learning，PMLR，2013:1310-1318.
[26]	LEI J. Key technology research and engineering practice of artificial intelligence in water environment monitoring［J］. Value Engineering，2019，38（22）:5. 雷健. 人工智能在水环境监测的关键技术研究与工程实践［J］. 价值工程，2019，38（22）:5.
[27]	KIM K J. Financial time series forecasting using support vector machines［J］. Neurocomputing，2003，55（1/2）:307-319.
[28]	CHENG Y，BI X，XU Y，et al. Artificial intelligence technologies in bioprocess:opportunities and challenges［J］. Bioresource Technology，2022:128451.
[29]	MASOOD A，AHMAD K. A review on emerging artificial intelligence（AI）techniques for air pollution forecasting:fundamentals，application and performance［J］. Journal of Cleaner Production，2021，322:129072.
[30]	SONG Y，ZHANG C，JIN X，et al. Spatial prediction of PM_2.5 concentration using hyper-parameter optimization XGBoost model in China［J］. Environmental Technology& Innovation，2023，32:103272.
[31]	CHEN T，GUESTRIN C. XGBoost［C］// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining-KDD，F，2016［C］.
[32]	FAN J，MA X，WU L，et al. Light gradient boosting machine:An efficient soft computing model for estimating daily reference evapotranspiration with local and external meteorological data［J］. Agricultural Water Management，2019，225:105758.
[33]	ROSENBLATT F. The perceptron:a probabilistic model for information storage and organization（1958）［M］. 2021:183-190.
[34]	WANG F，CHEN W，NIU L. An improved InVEST ecological service evaluation model based on BP neural network optimization［J］. Nature Environment and Pollution Technology，2020，19:991-1000.
[35]	BENEDETTI M，CESAROTTI V，INTRONA V，et al. Energy consumption control automation using artificial neural networks and adaptive algorithms:proposal of a new methodology and case study［J］. Applied Energy，2016，165:60-71.
[36]	NIKPEY H，ASSADI M，BREUHAUS P，et al. Experimental evaluation and ANN modeling of a recuperative micro gas turbine burning mixtures of natural gas and biogas［J］. Applied Energy，2014，117:30-41.
[37]	CHANNAPATTANA S，PAWAR A A，KAMBLE P G. Optimisation of operating parameters of DI-CI engine fueled with second generation bio-fuel and development of ANN based prediction model［J］. Applied Energy，2017，187:84-95.
[38]	CAUDILL M，BUTLER C. Understanding neural networks 1，basic networks［Z］. PressMIT，Cambridge，Mass. 1992
[39]	SHAMS S R，JAHANI A，KALANTARY S，et al. The evaluation on artificial neural networks（ANN）and multiple linear regressions（MLR）models for predicting SO₂ concentration［J］. Urban Climate，2021，37:100837.
[40]	LI Z，YANG Q，WANG L，et al. Application of RBFN network and GM（1，1）for groundwater level simulation［J］. Applied Water Science，2017，7（6）:3345-3353.
[41]	DRAGOVIĆ S. Artificial neural network modeling in environmental radioactivity studies:A review［J］. Science of the Total Environment，2022，847:157526.
[42]	VAHDATPOUR M S，SAJEDI H，RAMEZANI F. Air pollution forecasting from sky images with shallow and deep classifiers［J］. Earth Science Informatics，2018，11（3）:413-422.
[43]	DU J，VONG C M，CHEN C L P. Novel efficient RNN and LSTM-Like architectures:recurrent and gated broad learning systems and their applications for text classification［J］. IEEE Transactions on Cybernetics，2021，51（3）:1586-1597.
[44]	LI L，JIANG P，XU H，et al. Water quality prediction based on recurrent neural network and improved evidence theory:a case study of Qiantang River，China［J］. Environmental Science and Pollution Research，2019，26（19）:19879198-96.
[45]	HAQ M A，AHMED A，KHAN I，et al. Analysis of environmental factors using AI and ML methods［J］. Scientific Reports，2022，12（1）:13267.
[46]	ZHANG X，ZHANG Y，LU X，et al. Estimation of lower-stratosphere-to-troposphere ozone profile using long short-term memory（LSTM）［J］. Remote Sensing，2021，13（7）:1374.
[47]	MONIZ J R A，KRUEGER D. Nested LSTMS［C］// Asian Conference on Machine Learning，PMLR，2017:530-544.
[48]	JIN N，ZENG Y，YAN K，et al. Multivariate air quality forecasting with nested long short term memory neural network［J］. IEEE Transactions on Industrial Informatics，2021，17（12）:8514-8522.
[49]	SAKIZADEH M，MIRZAEI R，GHORBANI H. Support vector machine and artificial neural network to model soil pollution:a case study in Semnan Province，Iran［J］. Neural Computing and Applications，2017，28（11）:3229-3238.
[50]	CHEN S，GAO C，ZHANG P. Incorporation of data-mined knowledge into black-box SVM for interpretability［J］. ACM Trans Intell Syst Technol，2022，14（1）:1-22.
[51]	LEONG W C，KELANI R O，AHMAD Z. Prediction of air pollution index（API）using support vector machine（SVM）［J］. Journal of Environmental Chemical Engineering，2020，8（3）:103208.
[52]	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）:121612-121629.
[53]	HUANG M，MA Y，WAN J，et al. A sensor-software based on a genetic algorithm-based neural fuzzy system for modeling and simulating a wastewater treatment process［J］. Applied Soft Computing，2015，27:1-10.
[54]	RIYAZ R，PUSHPA P. Air quality prediction in smart cities:a fuzzy-logic based approach； proceedings of the 2018 International Conference on Computational Techniques，electronics and Mechanical Systems（CTEMS），F，2018［C］. IEEE.
[55]	LI Z S，LIANG X G，JIN Y K，et al. A comparative study on edictive effect of PM2.5 in Beijing based on tree models［J］. Environmental Engineering，2021，39（6）:106-113. 李志生，梁锡冠，金宇凯，等. 基于树模型的北京市PM_2.5预测效果对比分析［J］. 环境工程，2021，39（6）:106-113.
[56]	HUANG C T，FAN D P，LU J F，et al. Prediction of PM_2.5 and PM₁₀ concentration in Guangzhou based on deep learning model［J］. Environmental Engineering，2021，39（12）:135-140. 黄春桃，范东平，卢集富，等. 基于深度学习模型的广州市大气PM_2.5和PM₁₀浓度预测［J］. 环境工程，2021，39（12）:135-140.
[57]	Jing H，Wang Y. Research on urban air quality prediction based on ensemble learning of XGBoost［C］// E3S Web of Conferences. EDP Sciences，2020，165:02014.
[58]	ZHANG B，ZHANG Y，JIANG X. Feature selection for global tropospheric ozone prediction based on the BO-XGBoost-RFE algorithm［J］. Scientific Reports，2022，12（1）:9244.
[59]	KE G，MENG Q，FINLEY T，et al. Lightgbm:a highly efficient gradient boosting decision tree［J］. Advances in neural information processing systems，2017，30.
[60]	ZHANG W，WU C，TANG L，et al. Efficient time-variant reliability analysis of Bazimen landslide in the Three Gorges Reservoir Area using XGBoost and LightGBM algorithms［J］. Gondwana Research，2023，123:41-53.
[61]	XU T，CHEN B，REN Y，et al. Estimation of the ground-level SO₂ concentration in eastern China based on the LightGBM model and Himawari-8 TOAR［J］. Frontiers in Earth Science，2023，10:1037719.
[62]	JAMEI M，ALI M，MALIK A，et al. Air quality monitoring based on chemical and meteorological drivers:application of a novel data filtering-based hybridized deep learning model［J］. Journal of Cleaner Production，2022，374:134011.
[63]	CHI Y，ZHAN Y，WANG K，et al. Spatial distribution of multiple atmospheric pollutants in China from 2015 to 2020［J］. Remote Sensing，2023，15（24）:5705.
[64]	MINSKY M. Logical versus analogical or symbolic versus connectionist or neat versus scruffy［J］. AI Mag，1991，12:34-51.
[65]	ZENG H，XU Z W. Industrial sewage evaluation based on fuzzy neural network［J］. Procedia Environmental Sciences，2011，10:771-776.
[66]	SU H，ZHAO F. A novel substation fault diagnosis approach based on RS and ANN and ES［C］// Proceedings of the 2006 International Conference on Communications，Circuits and Systems，F，2006，IEEE.
[67]	WEN C H，VASSILIADIS C. Applying hybrid artificial intelligence techniques in wastewater treatment［J］. Engineering Applications of Artificial Intelligence，1998，11（6）:685-705.
[68]	TIWARI M K，BAJPAI S，DEWANGAN U. Prediction of industrial solid waste with ANFIS model and its comparison with ANN model-a case study of Durg-Bhilai twin city India［J］. International Journal of Engineering and Innovative Technology（IJEIT），2012，6（2）:192-201.
[69]	ABBASI M，HANANDEH A EL. Forecasting municipal solid waste generation using artificial intelligence modelling approaches［J］. Waste management，2016，56:13-22.
[70]	IRAJI M S，ABOUTALEBI M，SEYEDAGHAEE N R，et al. Students classification with adaptive neuro fuzzy［J］. International Journal of Modern Education and Computer Science，2012，4（7）:42.
[71]	CHANG L C，CHANG F J. Intelligent control for modelling of real-time reservoir operation［J］. Hydrological Processes，2001，15（9）:1621-1634.
[72]	MAHMOODI B N，HEYDARI A，NEMATOLLAHZADEH A. Removal of methylene blue via bioinspired catecholamine/starch superadsorbent and the efficiency prediction by response surface methodology and artificial neural network-particle swarm optimization［J］. Bioresource Technology，2019，294:122084.
[73]	EJIMOFOR M，EZEMAGU I，MENKITI M. RSM and ANN-GA modeling of colloidal particles removal from paint wastewater via coagulation method using modified Aguleri montmorillonite clay［J］. Current Research in Green and Sustainable Chemistry，2021，4:100164.
[74]	CHANG H H，PARK I，OH H R，et al. Development of a PM2.5 prediction model using a recurrent neural network algorithm for the Seoul metropolitan area，Republic of Korea［J］. Atmospheric Environment，2021，245.
[75]	DAI H，HUANG G，ZENG H，et al. PM_2.5 volatility prediction by XGBoost-MLP based on GARCH models［J］. Journal of Cleaner Production，2022，356:131898.
[76]	ALLAMSETTY S，MOHAPATRO S. Feasibility of artificial neural networks and fuzzy logic models for prediction of NOx concentrations in nonthermal plasma-treated diesel exhaust［J］. IEEE Transactions on Plasma Science，2019，47（5）:2637-2644.
[77]	CAKIR S，SITA M. Evaluating the performance of ANN in predicting the concentrations of ambient air pollutants in Nicosia［J］. Atmospheric Pollution Research，2020，11（12）:2327-2334.
[78]	TURIAS I J，GONZÁLEZ F J，MARTIN M L，et al. Prediction models of CO，SPM and SO₂ concentrations in the Campo de Gibraltar Region，Spain:a multiple comparison strategy［J］. Environmental monitoring and assessment，2008，143:131-146.
[79]	LIN M D，LIU P Y，HUANG C W，et al. The application of strategy based on LSTM for the short-term prediction of PM_2.5 in city［J］. Science of the Total Environment，2024，906:167892.
[80]	WANG G，GUO L，DONG X，et al. Artificial neural network prediction of air pollutants in Shenzhen city［J］. Chinese Journal of Environmental Engineering，2015，9（7）:3393-3399.
[81]	OSOWSKI S. Prediction of air pollution using LSTM［C］// Advances in Computational Intelligence:16th International Work-Conference on Artificial Neural Networks，IWANN 2021，Proceedings，Part II 16. Springer International Publishing，2021:208-219.
[82]	SIWEK K，OSOWSKI S. Data mining methods for prediction of air pollution［J］. International Journal of Applied Mathematics and Computer Science，2016，26（2）.
[83]	YILDIRIM Y，BAYRAMOGLU M. Adaptive neuro-fuzzy based modelling for prediction of air pollution daily levels in city of Zonguldak［J］. Chemosphere，2006，63（9）:1575-1582.
[84]	Ghasemi A，Amanollahi J. Integration of ANFIS model and forward selection method for air quality forecasting［J］. Air Quality，Atmosphere& Health，2019，12:59-72.
[85]	UNNIKRISHNAN R，MADHU G. Comparative study on the effects of meteorological and pollutant parameters on ANN modelling for prediction of SO₂［J］. SN Applied Sciences，2019，1:1-12.
[86]	RAGAB M G，ABDULKADIR S J，AZIZ N，et al. A novel one-dimensional CNN with exponential adaptive gradients for air pollution index prediction［J］. Sustainability，2020，12（23）:10090.
[87]	LIU H，WU H，LV X，et al. An intelligent hybrid model for air pollutant concentrations forecasting:a case of Beijing in China［J］. Sustainable Cities and Society，2019，47:101471.
[88]	ŞAHIN Ü A，BAYAT C，UÇAN O N. Application of cellular neural network（CNN）to the prediction of missing air pollutant data［J］. Atmospheric Research，2011，101（1）:314-326.
[89]	SHAD R，MESGARI M S，ABKAR A，et al. Predicting air pollution using fuzzy genetic linear membership kriging in GIS［J］. Computers，Environment and Urban Systems，2009，33（6）:472-481.
[90]	SINGH K，GUPTA S，KUMAR A，et al. Linear and nonlinear modeling approaches for urban air quality prediction［J］. The Science of the Total Environment，2012，426:244-255.
[91]	PRASAD K，GORAI A K，GOYAL P. Development of ANFIS models for air quality forecasting and input optimization for reducing the computational cost and time［J］. Atmospheric Environment，2016，128:246-262.
[92]	ASADOLLAHFARDI G，MEHDINEJAD M，PAM M，et al. Predicting carbon monoxide concentrations in the air of Pardis City，Iran，using an artificial neural network［J］. Environmental Quality Management，2016，26（1）:37-49.
[93]	KALATE A I，SALARI M，SHADMAN A. Statistical models for multi-step-ahead forecasting of fine particulate matter in urban areas［J］. Atmospheric Pollution Research，2019，10（3）:689-700.
[94]	FENG R，ZHENG H J，GAO H，et al. Recurrent Neural Network and random forest for analysis and accurate forecast of atmospheric pollutants:a case study in Hangzhou，China［J］. Journal of Cleaner Production，2019，231:1005-1015.
[95]	ALJANABI M，SHKOUKANI M，HIJJAWI M. Ground-level ozone prediction using machine learning techniques:a case study in Amman，Jordan［J］. International Journal of Automation and Computing，2020，17（5）:667-677.
[96]	SUN Q，ZHU Y，CHEN X，et al. A hybrid deep learning model with multi-source data for PM_2.5 concentration forecast［J］. Air Quality，Atmosphere& Health，2021，14（4）:503-513.
[97]	CORDOVA C H，PORTOCARRERO M N L，SALAS R，et al. Air quality assessment and pollution forecasting using artificial neural networks in Metropolitan Lima-Peru［J］. Scientific Reports，2021，11（1）:24232.
[98]	LYU J，YIN S，SHANG C，et al. Sensitivity analysis of isoprene and aerosol emission in a suburban plantation using long short-term memory model［J］. Urban Forestry& Urban Greening，2021，64:127303.
[99]	BRAZÁLEZ E，MACIÀ H，DÍAZ G，et al. FUME:an air quality decision support system for cities based on CEP technology and fuzzy logic［J］. Applied Soft Computing，2022:109536.