Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A hybrid deep learning technology for PM2.5 air quality forecasting.

Zhendong Zhang1, Yongkang Zeng1, Ke Yan2,3

  • 1Key Laboratory of Electromagnetic Wave Information Technology and Metrology of Zhejiang Province, College of Information Engineering, China Jiliang University, Hangzhou, 310018, China.

Environmental Science and Pollution Research International
|March 24, 2021
PubMed
Summary

Related Concept Videos

Linear Approximations01:23

Linear Approximations

For a differentiable function of two variables, linear approximation estimates values near a known point by replacing the curved surface with its tangent plane. Consider the function\begin{equation*}f(x,y)=x^2+3y^2\end{equation*}near the point (2, 1). The exact value at this point is f(2, 1) = 22 + 3(1)2 = 4 + 3 = 7.The linear approximation of f(x, y)) near (a, b) is\begin{equation*}L(x,y)=f(a,b)+f_x(a,b)(x-a)+f_y(a,b)(y-b)\end{equation*}First, compute the partial derivatives: fx(x, y) = 2x and...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Molecular characterization and pathogenicity of a HP-like PRRSV strain and a NADC30 recombinant strain.

Microbial pathogenesis·2026
Same author

Microbial ecology and flavor formation mechanisms of high-temperature Daqu in the Huang-Huai River basin and adjacent regions: A comparative study from eastern Henan, Jiaodong peninsula, and southern Anhui.

Food research international (Ottawa, Ont.)·2026
Same author

A malignant symbiosis: The neuro-metabolic symphony rewires the tumor microenvironment.

Neoplasia (New York, N.Y.)·2026
Same author

The Synergistic Refinement of Primary Si in Hypereutectic Al-Si Alloys Using Ce and Er Elements.

Materials (Basel, Switzerland)·2026
Same author

Spatial transcriptome analysis of the tea tender shoot sheds light on transcriptional regulation of characteristic metabolites.

Horticulture research·2026
Same author

The role and significance of the PI3K/Akt-mTOR signaling pathway in infectious diseases.

Frontiers in cellular and infection microbiology·2026
This summary is machine-generated.

A new hybrid deep learning model, Variational Mode Decomposition-Bidirectional Long Short-Term Memory (VMD-BiLSTM), accurately predicts fine particulate matter (PM2.5) air quality. This advanced VMD-BiLSTM approach outperforms existing methods for stable and reliable PM2.5 forecasting.

Area of Science:

  • Environmental Science
  • Data Science
  • Artificial Intelligence

Background:

  • Particulate Matter (PM2.5) concentration is a critical air quality indicator impacting public health, economics, and societal development.
  • Predicting volatile PM2.5 changes is challenging due to the inherent nonlinearity and instability of air quality data.
  • Existing forecasting models struggle to capture complex temporal dynamics effectively.

Purpose of the Study:

  • To develop and evaluate a novel hybrid deep learning model for accurate PM2.5 concentration prediction.
  • To enhance forecasting accuracy by decomposing complex time series data into manageable sub-signals.
  • To compare the proposed model's performance against established empirical mode decomposition (EMD) and other VMD-based methods.

Main Methods:

Keywords:
Air quality predictionBidirectional longDeep learningShort-term memory networkVariational mode decomposition

Related Experiment Videos

  • A hybrid model combining Variational Mode Decomposition (VMD) and Bidirectional Long Short-Term Memory (BiLSTM) networks was constructed.
  • VMD was utilized to decompose the original PM2.5 time series data into multiple intrinsic mode functions (sub-signals) in the frequency domain.
  • BiLSTM networks were independently applied to predict each decomposed sub-signal, leveraging their ability to capture long-range dependencies.

Main Results:

  • The proposed VMD-BiLSTM model demonstrated superior performance in PM2.5 forecasting accuracy compared to all evaluated models, including EMD-based and other VMD-based approaches.
  • Integrating VMD into forecasting models yielded significantly better results than integrating EMD.
  • The VMD-BiLSTM model proved to be the most stable forecasting method among all VMD-integrated models tested.

Conclusions:

  • The VMD-BiLSTM hybrid deep learning model offers a robust and accurate framework for predicting PM2.5 concentration changes.
  • The decomposition capability of VMD, combined with the predictive power of BiLSTM, effectively addresses the challenges of nonlinear air quality forecasting.
  • This study highlights the potential of hybrid deep learning approaches for improving environmental monitoring and public health protection strategies.