Loading [MathJax]/jax/output/SVG/jax.js
Research article Special Issues

Hybrid deep learning-based air pollution prediction and index classification using an optimization algorithm

  • Lockdowns were implemented in nearly all countries in the world in order to reduce the spread of COVID-19. The majority of the production activities like industries, transportation, and construction were restricted completely. This unprecedented stagnation of resident's consumption and industrial production has efficiently reduced air pollution emissions, providing typical and natural test sites to estimate the effects of human activity controlling on air pollution control and reduction. Air pollutants impose higher risks on the health of human beings and also damage the ecosystem. Previous research has used machine learning (ML) and statistical modeling to categorize and predict air pollution. This study developed a binary spring search optimization with hybrid deep learning (BSSO-HDL) for air pollution prediction and an air quality index (AQI) classification process during the pandemic. At the initial stage, the BSSO-HDL model pre-processes the actual air quality data and makes it compatible for further processing. In the presented BSSO-HDL model, an HDL-based air quality prediction and AQI classification model was applied in which the HDL was derived by the use of a convolutional neural network with an extreme learning machine (CNN-ELM) algorithm. To optimally modify the hyperparameter values of the BSSO-HDL model, the BSSO algorithm-based hyperparameter tuning procedure gets executed. The experimental outcome demonstrates the promising prediction classification performance of the BSSO-HDL model. This model, developed on the Python platform, was evaluated using the coefficient of determination R2, the mean absolute error (MAE), and the root mean squared error (RMSE) error measures. With an R2 of 0.922, RMSE of 15.422, and MAE of 10.029, the suggested BSSO-HDL technique outperforms established models such as XGBoost, support vector machines (SVM), random forest (RF), and the ensemble model (EM). This demonstrates its ability in providing precise and reliable AQI predictions.

    Citation: Sreenivasulu Kutala, Harshavardhan Awari, Sangeetha Velu, Arun Anthonisamy, Naga Jyothi Bathula, Syed Inthiyaz. Hybrid deep learning-based air pollution prediction and index classification using an optimization algorithm[J]. AIMS Environmental Science, 2024, 11(4): 551-575. doi: 10.3934/environsci.2024027

    Related Papers:

    [1] K. Wayne Forsythe, Cameron Hare, Amy J. Buckland, Richard R. Shaker, Joseph M. Aversa, Stephen J. Swales, Michael W. MacDonald . Assessing fine particulate matter concentrations and trends in southern Ontario, Canada, 2003–2012. AIMS Environmental Science, 2018, 5(1): 35-46. doi: 10.3934/environsci.2018.1.35
    [2] Suprava Ranjan Laha, Binod Kumar Pattanayak, Saumendra Pattnaik . Advancement of Environmental Monitoring System Using IoT and Sensor: A Comprehensive Analysis. AIMS Environmental Science, 2022, 9(6): 771-800. doi: 10.3934/environsci.2022044
    [3] Muhammad Rendana, Wan Mohd Razi Idris, Sahibin Abdul Rahim . Clustering analysis of PM2.5 concentrations in the South Sumatra Province, Indonesia, using the Merra-2 Satellite Application and Hierarchical Cluster Method. AIMS Environmental Science, 2022, 9(6): 754-770. doi: 10.3934/environsci.2022043
    [4] Shagufta Henna, Mohammad Amjath, Asif Yar . Time-generative AI-enabled temporal fusion transformer model for efficient air pollution sensor calibration in IIoT edge environments. AIMS Environmental Science, 2025, 12(3): 526-556. doi: 10.3934/environsci.2025024
    [5] Suwimon Kanchanasuta, Sirapong Sooktawee, Natthaya Bunplod, Aduldech Patpai, Nirun Piemyai, Ratchatawan Ketwang . Analysis of short-term air quality monitoring data in a coastal area. AIMS Environmental Science, 2021, 8(6): 517-531. doi: 10.3934/environsci.2021033
    [6] Carolyn Payus, Siti Irbah Anuar, Fuei Pien Chee, Muhammad Izzuddin Rumaling, Agoes Soegianto . 2019 Southeast Asia Transboundary Haze and its Influence on Particulate Matter Variations: A Case Study in Kota Kinabalu, Sabah. AIMS Environmental Science, 2023, 10(4): 547-558. doi: 10.3934/environsci.2023031
    [7] Winai Meesang, Erawan Baothong, Aphichat Srichat, Sawai Mattapha, Wiwat Kaensa, Pathomsorn Juthakanok, Wipaporn Kitisriworaphan, Kanda Saosoong . Effectiveness of the genus Riccia (Marchantiophyta: Ricciaceae) as a biofilter for particulate matter adsorption from air pollution. AIMS Environmental Science, 2023, 10(1): 157-177. doi: 10.3934/environsci.2023009
    [8] Joanna Faber, Krzysztof Brodzik . Air quality inside passenger cars. AIMS Environmental Science, 2017, 4(1): 112-133. doi: 10.3934/environsci.2017.1.112
    [9] Pratik Vinayak Jadhav, Sairam V. A, Siddharth Sonkavade, Shivali Amit Wagle, Preksha Pareek, Ketan Kotecha, Tanupriya Choudhury . A multi-task model for failure identification and GPS assessment in metro trains. AIMS Environmental Science, 2024, 11(6): 960-986. doi: 10.3934/environsci.2024048
    [10] Anna Mainka, Barbara Kozielska . Assessment of the BTEX concentrations and health risk in urban nursery schools in Gliwice, Poland. AIMS Environmental Science, 2016, 3(4): 858-870. doi: 10.3934/environsci.2016.4.858
  • Lockdowns were implemented in nearly all countries in the world in order to reduce the spread of COVID-19. The majority of the production activities like industries, transportation, and construction were restricted completely. This unprecedented stagnation of resident's consumption and industrial production has efficiently reduced air pollution emissions, providing typical and natural test sites to estimate the effects of human activity controlling on air pollution control and reduction. Air pollutants impose higher risks on the health of human beings and also damage the ecosystem. Previous research has used machine learning (ML) and statistical modeling to categorize and predict air pollution. This study developed a binary spring search optimization with hybrid deep learning (BSSO-HDL) for air pollution prediction and an air quality index (AQI) classification process during the pandemic. At the initial stage, the BSSO-HDL model pre-processes the actual air quality data and makes it compatible for further processing. In the presented BSSO-HDL model, an HDL-based air quality prediction and AQI classification model was applied in which the HDL was derived by the use of a convolutional neural network with an extreme learning machine (CNN-ELM) algorithm. To optimally modify the hyperparameter values of the BSSO-HDL model, the BSSO algorithm-based hyperparameter tuning procedure gets executed. The experimental outcome demonstrates the promising prediction classification performance of the BSSO-HDL model. This model, developed on the Python platform, was evaluated using the coefficient of determination R2, the mean absolute error (MAE), and the root mean squared error (RMSE) error measures. With an R2 of 0.922, RMSE of 15.422, and MAE of 10.029, the suggested BSSO-HDL technique outperforms established models such as XGBoost, support vector machines (SVM), random forest (RF), and the ensemble model (EM). This demonstrates its ability in providing precise and reliable AQI predictions.



    The method of fundamental solutions (MFS) [1,2] is a simple, accurate and efficient boundary-type meshfree approach for the solution of partial differential equations, which uses the fundamental solution of a differential operator as a basis function. Chen et al. [3,4] demonstrated the equivalence between the MFS and the Trefftz collocation method [5] under certain conditions. During the last decades, the MFS has been widely applied to various fields in applied mathematics and mechanics, such as scattering and radiation problems [6], inverse problems [7], non-linear problems [8], and fractal derivative models [9,10]. For more details and applications, the readers are referred to Refs. [11,12,13] and the references therein. In the MFS, the approximation solution is assumed as a linear combination of fundamental solutions. In order to avoid the source singularity of the fundamental solution, this method requires a fictitious boundary outside the computational domain, and places the source points on it. The distance between the fictitious boundary and the real boundary has a certain influence on the calculation accuracy [14,15,16]. To address this issue, many scholars proposed improved methods, such as the singular boundary method [17,18], modified method of fundamental solutions [19], the non-singular method of fundamental solutions [20]. Due to the characteristic of dense matrix, the traditional and improved MFS with "global" discretization is difficult to apply in the large-scale problems with complicated geometries.

    Recently, Fan and Chen et al. [21] proposed an improved version of traditional MFS, known as the localized method of fundamental solutions (LMFS), for solving Laplace and biharmonic equations. The LMFS is essentially a localized semi-analytical meshless collocation scheme, and overcomes the limitation of traditional method in applications of complex geometry and large-scale problems. Gu and Liu et al. [22,23,24,25] extended to the LMFS to elasticity, heat conduction and inhomogeneous elliptic problems. Qu et al. [26,27,28] applied the method to the interior acoustic and bending analysis. Wang et al. [29,30,31,32] developed the localized space-time method of fundamental solutions, and resolved the inverse problems. Li et al. [33] used this method to simulate 2D harmonic elastic wave problems. Liu et al. [34] combined the LMFS and the Crank-Nicolson time-stepping technology to address the transient convection-diffusion-reaction equations. Li, Qu and Wang et al. [35,36,37] given the error analysis of the LMFS, and proposed a augmented moving least squares approximation. Like the element-free Galerkin method [38,39] being successfully applied to a large number of partial differential equations, the proposed LMFS belongs to the domain-type meshless method. Unlike the former, the latter is a semi-analytical meshless collocation method with strong form, in which the fundamental solutions are unavoidable. In the LMFS, the whole computational domain is firstly divided into a set of overlapping local subdomains whose boundary could be a circle (for 2D problems) and/or a sphere (for 3D problems). In each of the local subdomain, the classical MFS formulation and the moving least square (MLS) method are applied to the local approximation of variables. This method remains the high accuracy of the traditional MFS, and simultaneously produces a sparse system of linear algebraic equations. Inspired by the LMFS, recently, some new local meshless methods [40,41,42,43,44] have also been proposed.

    No methodology is perfect, although the LMFS has achieved varying degrees of success in various fields, it still faces some thorny problems to be solved. As a local semi-analytical meshless technique, the LMFS needs to configure nodes in the whole computational domain and to select the supporting nodes in each local subdomain, which likes the generalized finite difference method [45,46,47]. The node spacing (related to the total number of nodes), the number of supporting nodes and the relationship between them have the certain influence on the accuracy. Fu et al. [48] discussed the selection algorithm of supporting nodes in the generalized finite difference method. However, there are few reports in the literature coordinating the node spacing and the number of supporting nodes in the LMFS.

    In this paper, we propose an effective empirical formula to determine the node spacing and the number of supporting nodes in the LMFS for solving 2D potential problems with Dirichlet boundary condition. A reasonable number of supporting nodes can be directly given according to the size of node spacing. The proposed methodology is applicable to both regular and irregular distribution of nodes in arbitrary domain. This study consummates the LMFS and lays a foundation for simulating various engineering problems accurately and stably.

    The rest of paper is organized as follows. In Section 2, we give the governing equation and boundary condition of 2D potential problems, and provide the numerical implementation of LMFS. Section 3 introduces the empirical formula of the node spacing and the number of supporting nodes in the LMFS. Section 4 investigates two numerical examples with irregular domain and nodes to illustrate the accuracy and reliability of the empirical formula. Section 5 summarizes some conclusions.

    Considering the following 2D potential problem with the Dirichlet boundary condition:

    2u(x,y)=0,(x,y)Ω, (1)
    u=ˉu(x,y),(x,y)Γ, (2)

    where 2 is the 2D Laplacian, u(x,y) the unknown variable, Ω the computational domain, Γ=Ω the boundary of domain, and ˉu(x,y) the given function.

    According to the ideas of the LMFS, N=ni+nb nodes xi (i=1,2,...,N) should be distributed inside the considered domain Ω and along its boundary Γ, where ni is the number of interior nodes, nb is the number of boundary nodes. Consider an arbitrary node x(0) (called the central node), we can find m supporting nodes x(i) (i=1,2,...,m) around the central node x(0) (see Figure 1 (a)). At the same time, the local subdomain Ωs can also be defined. Subsequently, the MFS formulation is implemented for the local subdomain. For this purpose, an artificial boundary Ωs is selected at a certain distance from the boundary of local subdomain, as shown in Figure 1 (b). On the artificial boundary, M uniformly distributed source points are specified.

    Figure 1.  Schematic diagram of the LMFS: (a) nodal distribution and (b) local subdomain.

    For each node in the local subdomain Ωs, the following MFS formulation of numerical solution should be hold

    u(x(i))=Mj=1αjG(x(i),s(j)),x(i)Ωs,i=0,1,...,m, (3)

    or for brevity

    u(i)=Mj=1αjGij=G(i)α,x(i)Ωs,i=0,1,...,m, (4)

    where (x(i))mi=0 are the m+1 nodes in the subdomains Ωs, (s(j))Mj=1 denote M fictitious source points, α=(α1,α2,,αM)T represent the unknown coefficient vector, Gij=G(x(i),s(j)) are the fundamental solutions expressed as

    G(x(i),s(j))=12πlnx(i),s(j)2. (5)

    It should be pointed out that the artificial boundary is a circle centered at x(0) and with radius Rs (see Figure 1 (b)), here Rs is a parameter that should be manually fixed by the user.

    In each local subdomain, we determine the unknown coefficients (αj)Mj=1 in Eq (3) or (4) by the moving least square (MLS) method, we can define a residual function as follows

    B(α)=mi=0[G(i)αu(i)]2ω(i), (6)

    in which ω(i) represents the weighting function related with node x(i) and is introduced as

    ω(i)=exp[(di/h)2]exp[(dmax/h)2]1exp[(dmax/h)2],i=0,1,,m, (7)

    where di=x(i)x(0)2 denotes the distance of the central node and its ith supporting node, dmax is the size of the local subdomain (the maximum value of distances between x(0) and m supporting nodes, i.e., dmax=maxi=1,2,...,m(di)), and h=1.

    Based on the MLS approximation, we can get the coefficient vector α=(α1,α2,,αM)T by minimizing B(α) with respect to α as follows

    B(α)αj=0,j=1,2,,M, (8)

    Then, a linear system can be formed

    Aα=b, (9)

    where

    A=[mi=0G2i1ω(i)mi=0Gi1Gi2ω(i)mi=0Gi1Gi3ω(i)mi=0Gi1GiMω(i)mi=0G2i2ω(i)mi=0Gi2Gi3ω(i)mi=0Gi2GiMω(i)mi=0G2i3ω(i)mi=0Gi3GiMω(i)SYMmi=0G2iMω(i)],b=[mi=0Gi1ω(i)u(i)mi=0Gi2ω(i)u(i)mi=0Gi3ω(i)u(i)mi=0GiMω(i)u(i)]. (10)

    The vector b in Eq (10) can be further expressed as

    b=[G01ω(0)G11ω(1)Gm1ω(m)G02ω(0)G12ω(1)Gm2ω(m)G0Mω(0)G1Mω(1)GmMω(m)](u(0)u(1)u(m))=B(u(0)u(1)u(m)). (11)

    According to Eqs (9)–(11), the unknown coefficients α=(α1,α2,,αM)T can now be calculated as

    α=(α1α2αM)=A1B(u(0)u(1)u(m)). (12)

    To ensure the regularity of matrix A in Eq (12), m+1 should be greater than M. For simplicity, we fixed m+1=2M in the computations. It should be pointed out that the matrix A with a small size given in Eq (10) is well-conditioned. In this study, the MATLAB routine "AB" is used to calculate A1B in order to avoid the troublesome matrix inversion.

    Substituting Eq (12) into Eq (4) as i=0, the numerical solution at central node x(0) is expressed as

    u(0)=G(0)α=G(0)A1B(u(0)u(1)u(m))=mj=0c(j)u(j), (13)

    or

    u(0)mj=0c(j)u(j)=0, (14)

    in which (c(j))mj=0 are coefficients which can be calculated by G(0)A1B.

    Now we can obtained a final linear algebraic equation system of the LMFS according to the governing equation and boundary condition. At interior nodes, the physical quantities must satisfy the governing equation, namely,

    uimj=0c(j)iu(j)=0,i=1,2,,ni, (15)

    where subscript i of c(j)i is used to distinguish the coefficients for different interior nodes. At boundary nodes with Dirichlet boundary condition, we have

    ui=ˉui,i=ni+1,ni+2,,ni+nd. (16)

    Using the given boundary data and combing Eqs (15) and (16), we have the following sparse system of linear algebraic equations

    CU=f, (17)

    where CN×N represents the coefficient matrix, U=(u1,u2,,uN)T denotes the undetermined vector of variables at all nodes, and fN×1 is a known vector composed by given boundary condition and zero vector. It should be pointed out that the system given in Eq (17) is well-conditioned, and standard solvers can be used to obtain its solution. In this study, the MATLAB routine "U=Cf" is used to solve this system of equation. After solving Eq (17), the approximated solutions at all nodes can be acquired.

    In this section, three different analytical solutions are provided to summarize the relationship between the number of supporting nodes (m) and the node spacing (Δh). Without loss of generality, the node spacing is defined by

    Δh=max1iNmin1jN|xixj|. (18)

    Noted that the node spacing is equivalent to the total number of nodes, and is suitable for the arbitrary distribution of nodes including regular and irregular distributions. In the investigation, the equidistant nodes are used, thus the node spacing Δh is the distance between two adjacent nodes. In all calculations, the artificial radius is fixed with Rs=1.5, and the numerical results are calculated on a computer equipped with i5-5200 CPU@2.20GHz and 4GB memory. To estimate the accuracy of the present scheme, we adopt the root-mean-square error (RMSE) defined by

    RMSE=1NtotalNtotalk(unum(xk)uexa(xk))2, (19)

    where unum(xk) and uexa(xk) are the numerical and analytical solutions at kth test points respectively, Ntotal is the total number of tested points, which refers to all nodes unless otherwise specified.

    As shown in Figure 2, a rectangular with equidistant nodes is considered. To obtain a relatively universal formula, three different exact solutions are employed, as shown below:

    u(x,y)=x2y2, (20)
    u(x,y)=sin(x)ey, (21)
    u(x,y)=cos(x)cosh(y)+sin(x)sinh(y). (22)
    Figure 2.  Computational domain and nodal distribution.

    In order to numerically analyze the relationship between m and Δh, we choose different m and Δh, and draw the RMSE profiles obtained by the LMFS. For the analytical solution in Eq (20), the fitting curve can be formulated as m=aΔh(b) ( denotes the round up operation), where 2.057<a<3.759 and 0.305<b<0.1509. Figure 3 plots the case with a=3 and b=0.22. For the analytical solution in Eq (21), the fitting curve can be formulated as m=aΔh(b), where 2.221<a<3.314 and 0.2817<b<0.1765. Figure 4 plots the case with a=2.76 and b=0.23. For the analytical solution in Eq (22), the fitting curve can be formulated as m=aΔh(b), where 2.61<a<3.153 and 0.2437<b<0.1937. Figure 5 plots the case with a=2.85 and b=0.22.

    Figure 3.  Distribution of RMSEs under different m and Δh.
    Figure 4.  Distribution of RMSEs under different m and Δh.
    Figure 5.  Distribution of RMSEs under different m and Δh.

    By considering the above three cases for different analytical solutions, the following simple empirical formula can be carefully concluded:

    m=3Δh(0.22). (23)

    From Eq (23), we can easily determine the number of supporting nodes by the node spacing. In the next section, two complex examples will be provided to verify the present expression.

    This section tests two numerical examples with complex boundaries. The following three different analytical solutions are used to to carefully validate the reliability of the empirical formula.

    u=sin(x)ey, (24)
    u=cos(x)cosh(y)+sin(x)sinh(y), (25)
    u=cos(x)cosh(y)+x2y2+2x+3y+1. (26)

    Example 1:

    A gear-shaped domain is considered as shown in Figure 6. The LMFS uses 1334 nodes including 1134 internal nodes and 200 boundary nodes generated from the MATLAB codes, The node spacing is Δh=0.0667m calculated from Eq (18). It can be known from the empirical formula (23) that the number of supporting nodes should be m6, when the numerical error remains RMSE1.0e02.

    Figure 6.  Distribution of nodes on the gear-shaped model.

    Figure 7 illustrates the RMSEs of numerical solutions at all nodes with respect to the number of supporting nodes for the different analytical solutions, where A, B and C represent the analytical solutions (24), (25) and (26), respectively. As can be seen, the numerical results for different types of exact solutions converge with increasing number of supporting nodes. More importantly, it can be observed from Figure 7 that RMSE1.0e02 when m6, indicating the reliability and accuracy of the proposed empirical formula.

    Figure 7.  Error curves of the LMFS under different number of supporting nodes.

    Example 2:

    In the second example, we consider a car cavity model. Figure 8 shows the problem geometry and the distribution of nodes. The total number of nodes is N=11558, including 11212 internal nodes and 346 boundary nodes. Nodes in this model are derived from the HyperMesh software. The node spacing is Δh=0.02m calculated from Eq (18). It should be pointed out that According to the proposed empirical formula (23), the number of supporting nodes needs to meet m8 when the numerical error remains RMSE1.0e02.

    Figure 8.  Geometry of the problem and the distribution of nodes.

    Figure 9 depicts error curves of the LMFS with the increase of the number of supporting nodes, under deferent tested solutions. Despite the complicated domain and the irregular-distributed nodes, the results in Figure 9 are in good agreement with our empirical formula. The above numerical results fully demonstrate the reliability of the present methodology. It is worth mentioning that a large number of numerical experiments have been performed with the empirical formula in Eq (23), and all experiments observe the similar performance.

    Figure 9.  Error curves of the LMFS under different number of supporting nodes.

    In this paper, a simple, accurate and effective empirical formula is proposed to choose the number of supporting nodes. As a local collocation method, the number of supporting nodes has an important impact on the accuracy of the LMFS. This study given the direct relationship between the number of supporting nodes and the node spacing. By using the proposed formula, a reasonable number of supporting nodes can be determined according to the node spacing. Numerical results demonstrate the validity of the developed formula. The proposed empirical formula is beneficial to accuracy, simplicity and university for selecting the number of supporting nodes in the LMFS. This paper perfected the theory of the LMFS, and provided a quantitative selection strategy of method parameters.

    It should be noted that the present study focuses on the 2D homogeneous linear potential problems with Dirichlet boundary condition. The proposed formula can not be directly applied to the 2D cases with Neumann boundary and 3D cases with Dirichlet or Neumann boundary condition. In addition, the LMFS depends on the fundamental solutions of governing equations. For the nonhomogeneous and/or nonlinear problems without the fundamental solutions, the appropriate auxiliary technologies should be introduced in the LMFS, and then the corresponding formula still need for further study. The subsequent works will be emphasized on these issues, according to the idea developed in this paper.

    The work described in this paper was supported by the Natural Science Foundation of Shandong Province of China (No. ZR2019BA008), the China Postdoctoral Science Foundation (No. 2019M652315), the National Natural Science Foundation of China (Nos.11802151, 11802165, 11872220), and Qingdao People's Livelihood Science and Technology Project (No. 19-6-1-88-nsh).

    The authors declare that they have no conflicts of interest to report regarding the present study.



    [1] Rahman M M, Paul K C, Hossain M A, et al. (2021) Machine Learning on the COVID-19 Pandemic, Human Mobility and Air Quality: A Review. IEEE Access 9: 72420–72450. https://doi.org/10.1109/ACCESS.2021.3079121 doi: 10.1109/ACCESS.2021.3079121
    [2] Xing X, Xiong Y, Yang R, et al. (2021) Predicting the effect of confinement on the COVID-19 spread using machine learning enriched with satellite air pollution observations. Proc Natl Acad Sci 118: 33. https://doi.org/10.1073/pnas.2109098118 doi: 10.1073/pnas.2109098118
    [3] Sethi J K, Mittal M (2020) Monitoring the Impact of Air Quality on the COVID-19 Fatalities in Delhi, India: Using Machine Learning Techniques. Disaster Med Public Health Prep 6: 604-611. https://doi.org/10.1017/dmp.2020.372 doi: 10.1017/dmp.2020.372
    [4] Yang J, Wen Y, Wang Y, et al. (2021) From COVID-19 to future electrification: Assessing traffic impacts on air quality by a machine-learning model. P Nati A Sci 118: e2102705118. https://doi.org/10.1073/pnas.2102705118 doi: 10.1073/pnas.2102705118
    [5] Rybarczyk Y, Zalakeviciute R (2021) Assessing the COVID‐19 Impact on Air Quality: A Machine Learning Approach. Geophysl Res Lett 48: e2020GL091202. https://doi.org/10.1029/2020GL091202 doi: 10.1029/2020GL091202
    [6] Liu H, Yue F, Xie Z (2022) Quantify the role of anthropogenic emission and meteorology on air pollution using machine learning approach: A case study of PM2.5 during the COVID-19 outbreak in Hubei Province, China. Environ Pollut 300: 118932. https://doi.org/10.1016/j.envpol.2022.118932 doi: 10.1016/j.envpol.2022.118932
    [7] Gatti R C, Velichevskaya A, Tateo A, et al. (2020) Machine learning reveals that prolonged exposure to air pollution is associated with SARS-CoV-2 mortality and infectivity in Italy. Environ Pollut 267: 115471. https://doi.org/10.1016/j.envpol.2020.115471 doi: 10.1016/j.envpol.2020.115471
    [8] Gao M, Yang H, Xiao Q, et al. (2022) COVID-19 lockdowns and air quality: Evidence from grey spatiotemporal forecasts. Socio-Econ Plan Sci 83: 101228. https://doi.org/10.1016/j.seps.2022.101228 doi: 10.1016/j.seps.2022.101228
    [9] Wijnands J S, Nice K A, Seneviratne S, et al. (2022) The impact of the COVID-19 pandemic on air pollution: A global assessment using machine learning techniques. Atmos Pollut Res 13: 101438. https://doi.org/10.1016/j.apr.2022.101438 doi: 10.1016/j.apr.2022.101438
    [10] Wibowo F W (2021) Prediction of air quality in Jakarta during the COVID-19 outbreak using long short-term memory machine learning. IOP Conference Series: Earth and Environmental Science 704: 012046. https://doi.org/10.1088/1755-1315/704/1/012046 doi: 10.1088/1755-1315/704/1/012046
    [11] Stephan T, Al-Turjman F, Ravishankar M, et al. (2022) Machine learning analysis on the impacts of COVID-19 on India's renewable energy transitions and air quality. Environ Sci Pollut Res 29: 79443–79465. doi: 10.1007/s11356-022-20997-2. https://doi.org/10.1007/s11356-022-20997-2 doi: 10.1007/s11356-022-20997-2
    [12] Li G, Tang Y, Yang H (2022) A new hybrid prediction model of air quality index based on secondary decomposition and improved kernel extreme learning machine. Chemosphere 305: 135348. https://doi.org/10.1016/j.chemosphere.2022.135348 doi: 10.1016/j.chemosphere.2022.135348
    [13] Yang H, Zhang Y, Li G (2023) Air quality index prediction using a new hybrid model considering multiple influencing factors: A case study in China. Atmos Pollut Res 14: 1016777. https://doi.org/10.1016/j.apr.2023.101677 doi: 10.1016/j.apr.2023.101677
    [14] Sassi M S H, Fourati L C (2021) Deep Learning and Augmented Reality for IoT-based Air Quality Monitoring and Prediction System. IEEE 2021. https://doi.org/10.1109/ISNCC52172.2021.9615639
    [15] Shahne M Z, Sezavar A, Najibi F (2022) A hybrid deep learning model to forecast air quality data based on COVID-19 outbreak in Mashhad, Iran. Ann Civ Environ Eng 6: 019–025. https://doi.org/10.29328/journal.acee.1001035 doi: 10.29328/journal.acee.1001035
    [16] Tsan Y T, Kristiani E, Liu P Y, et al. (2022) In the Seeking of Association between Air Pollutant and COVID-19 Confirmed Cases Using Deep Learning. Int J Environl Res Pub He 19: 6373. https://doi.org/10.3390/ijerph19116373 doi: 10.3390/ijerph19116373
    [17] Lovrić M, Pavlović K, Vuković M, et al. (2021) Understanding the true effects of the COVID-19 lockdown on air pollution by means of machine learning. Environ Pollut 274: 115900. https://doi.org/10.1016/j.envpol.2020.115900 doi: 10.1016/j.envpol.2020.115900
    [18] Tyagi A, Gaur L, Singh G, et al. (2022) Air Quality Index (AQI) Using Time Series Modelling During COVID Pandemic. Lect Notes Electr Eng 2022: 441–452. https://doi.org/10.1007/978-981-16-8546-0_36 doi: 10.1007/978-981-16-8546-0_36
    [19] Maltare N N, Vahora S (2023) Air quality index prediction using machine learning for Ahmedabad city. Digital. Chemical. Engineering 7: 100093. https://doi.org/10.1016/j.dche.2023.100093 doi: 10.1016/j.dche.2023.100093
    [20] Xu J, Wang S, Ying N, et al. (2023) Dynamic graph neural network with adaptive edge attributes for air quality prediction: A case study in China. Heliyon 9: 17746. https://doi.org/10.1016/j.heliyon.2023.e17746 doi: 10.1016/j.heliyon.2023.e17746
    [21] Ghoneim A, Muhammad G, Hossain M S (2020) Cervical cancer classification using convolutional neural networks and extreme learning machines. Future Gener Comp Sy 102: 643–649. https://doi.org/10.1016/j.future.2019.09.015 doi: 10.1016/j.future.2019.09.015
    [22] Dehghani M, Montazeri Z, Dehghani A, et al. (2021) Binary Spring Search Algorithm for Solving Various Optimization Problems. Appl Sci 11: 1286. https://doi.org/10.3390/app11031286 doi: 10.3390/app11031286
    [23] Kamalraj R, Neelakandan S, Kumar M R, et al. (2021) Interpretable filter based convolutional neural network (IF-CNN) for glucose prediction and classification using PD-SS algorithm. Measurement 183: 109804. https://doi.org/10.1016/j.measurement.2021.109804 doi: 10.1016/j.measurement.2021.109804
    [24] Kavitha T, Mathai P P, Karthikeyan C, et al. (2021) Deep Learning Based Capsule Neural Network Model for Breast Cancer Diagnosis Using Mammogram Images. Interdiscip Sci 2021: 1-17. https://doi.org/10.1007/s12539-021-00467-y doi: 10.1007/s12539-021-00467-y
    [25] Reshma G, Al-Atroshi C, Nassa V K, et al. (2022) Deep Learning-Based Skin Lesion Diagnosis Model Using Dermoscopic Images. Intell Autom Soft Co 31: 621–634. https://doi.org/10.32604/iasc.2022.019117 doi: 10.32604/iasc.2022.019117
    [26] Harshavardhan A, Boyapati P, Neelakandan S, et al. (2022) LSGDM with Biogeography-Based Optimization (BBO) Model for Healthcare Applications. J Healthc Eng 2022: 1–11. https://doi.org/10.1155/2022/2170839 doi: 10.1155/2022/2170839
    [27] Neelakandan S, Beulah J R, Prathiba L, et al. (2022) Blockchain with deep learning-enabled secure healthcare data transmission and diagnostic model. Int J Model Simul Sc 13: 2241006. https://doi.org/10.1142/S1793962322410069 doi: 10.1142/S1793962322410069
    [28] Mao W, Wang W, Jiao L, et al. (2020) Modeling air quality prediction using a deep learning approach: Method optimization and evaluation. Sustain Cities Soc 65: 102567. https://doi.org/10.1016/j.scs.2020.102567 doi: 10.1016/j.scs.2020.102567
    [29] Jurado X, Reiminger N, Benmoussa M, et al. (2022). Deep learning methods evaluation to predict air quality based on Computational Fluid Dynamics. Expert System Applications 203: 117294. https://doi.org/10.1016/j.eswa.2022.117294 doi: 10.1016/j.eswa.2022.117294
  • This article has been cited by:

    1. Bingrui Ju, Wenzhen Qu, Yan Gu, Boundary Element Analysis for Mode III Crack Problems of Thin-Walled Structures from Micro- to Nano-Scales, 2023, 136, 1526-1506, 2677, 10.32604/cmes.2023.025886
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1872) PDF downloads(131) Cited by(0)

Figures and Tables

Figures(15)  /  Tables(7)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog