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 Missing Data Imputation Method for Gas Time Series Based on Spatio-Temporal Graph Attention Network-Echo State

Jian Yang1, Kai Qin1,2, Jinjiao Ye1,2

  • 1China Coal Research Institute, Beijing 100013, China.

Sensors (Basel, Switzerland)
|May 27, 2026
PubMed
Summary

Related Concept Videos

Time-Series Graph00:54

Time-Series Graph

A time-series graph is a line graph with repeated measurements taken at successive intervals of time. It is also called a time series chart. To construct a time-series graph, one must look at both pieces of a paired data set. The horizontal axis is used to plot the time increments, and the vertical axis is used to plot the values of the variable that one is measuring. By using the axes in this way, each point on the graph will correspond to time and a measured quantity. The points on the graph...

You might also read

Related Articles

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

Sort by
Same author

Reconstruction of Tectonic Stress Magnitude of the Yanshanian Period in the Pingdingshan Mining Area by Using the Microstructural Estimation Method.

ACS omega·2026
Same author

Study on Ultrasonic Modification of CH<sub>4</sub>/CO<sub>2</sub> Competitive Adsorption Behavior in Coal: A Pore Wall Molecular Structure Perspective.

Langmuir : the ACS journal of surfaces and colloids·2026
Same author

High-Precision Coal Mine Microseismic P-Wave Arrival Picking via Physics-Constrained Deep Learning.

Sensors (Basel, Switzerland)·2025
Same author

Abnormal Pressure Event Recognition and Dynamic Prediction Method for Fully Mechanized Mining Working Face Based on GRU-AM.

Sensors (Basel, Switzerland)·2025
Same author

Design of Low-Frequency Extended Signal Conditioning Circuit for Coal Mine Geophone.

Sensors (Basel, Switzerland)·2025
Same author

Environmental-Friendly, Long-Lastingly Moist Phase Change Gel with Tunable Cross-Linking Time for Effective Borehole Sealing.

Langmuir : the ACS journal of surfaces and colloids·2024

This study introduces a novel Spatio-Temporal Graph Attention Network-Echo State Network (ST-GAT-ESN) for accurate coal mine gas data imputation. The method effectively addresses missing data, improving mine safety analysis and disaster early warning systems.

Area of Science:

  • Data Science
  • Artificial Intelligence
  • Mining Engineering

Background:

  • Coal-mine-gas-monitoring data is crucial for mine safety but suffers from missing values due to harsh environments.
  • Existing imputation methods struggle with nonlinear spatiotemporal correlations and long-range dependencies in gas-monitoring time-series data.
  • Accurate imputation ensures data continuity, enhances disaster early warning reliability, and improves safety analysis.

Purpose of the Study:

  • To propose an advanced missing data imputation method for coal mine gas time-series data.
  • To address random and segmented missing data issues prevalent in underground monitoring.
  • To improve the accuracy and robustness of gas data imputation for enhanced mine safety.

Main Methods:

  • Utilized a Gated Recurrent Unit (GRU) for temporal feature extraction.
Keywords:
Echo State NetworkGraph Attention NetworkST-GAT-ESN modelmissing-value imputationspatio-temporal attention model

Related Experiment Videos

  • Employed a Graph Attention Network (GAT) to model monitoring points as nodes, using airflow relationships for an adjacency matrix to capture spatiotemporal dependencies.
  • Developed a dual-channel Echo State Network (ESN) with ridge regression for accurate imputation of missing values by fitting nonlinear trends.

    • Main Results:

    • The proposed Spatio-Temporal Graph Attention Network-Echo State Network (ST-GAT-ESN) method demonstrated optimal imputation performance for both random and segmented missing data (5-50% missing rate).
    • Achieved significant reductions in Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) by 30-80% compared to benchmark models.
    • The imputation curve closely matched the ground-truth curve, even at a 50% segmented missing rate, indicating high accuracy and stability.

    Conclusions:

    • The ST-GAT-ESN model effectively handles complex missing patterns through spatiotemporal collaborative modeling and a dual-channel fusion mechanism.
    • This provides a high-precision, stable technical solution for ensuring the integrity of coal-mine-gas-monitoring data.
    • Offers valuable theoretical references and engineering insights for missing-value processing in industrial time-series monitoring data.