Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Oct 3;15(1):34472.
doi: 10.1038/s41598-025-19063-x.

Fast and robust mixed gas identification and recognition using tree-based machine learning and sensor array response

Ghazala Ansari  1 , Rupali Singh  2 , Sachin Kumar  3 , Naglaa F Soliman  4
Affiliations

Fast and robust mixed gas identification and recognition using tree-based machine learning and sensor array response

Ghazala Ansari et al. Sci Rep. .

Abstract

The identification of gas mixtures is critical in chemical engineering, food safety, environmental science, and medical applications. This study employs a four-sensor array to classify ethylene-methane and ethylene-carbon monoxide (CO) mixtures, with concentrations ranging from 0 to 20 ppm for ethylene, 0-600 ppm for CO, and 0-300 ppm for methane. To enhance robustness, mean sensor response over time is used to mitigate noise, ensuring high classification accuracy. Each sample is evaluated using sixteen features, including temporal dynamics (e.g., rise time) and statistical metrics (e.g., baseline variance). Tree-based machine learning models-Decision Tree (DT), Random Forest (RF), and Extra Trees (ET)-are developed for gas classification, significantly reducing training set size (to 60%) and prediction time. The proposed ET model achieves superior classification accuracy (99.15%) compared to RF (95.86%) and DT (93.89%) on the UCI dataset, offering an efficient and accurate method for experimental gas identification.

PubMed Disclaimer

Conflict of interest statement

Declarations. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Gas sensors and their applications in different fields.
Fig. 2
Fig. 2
(a) Dynamic response curves for gas sensors, (b) Gas mixtures with varied concentrations under dynamic conditions, (c) ML model with tree-based cross-validation based on k-folds (k = 10), (d) Tree-based ML models.
Fig. 3
Fig. 3
16 chemical gas sensor array response for the mixture of ethylene-CO.
Fig. 4
Fig. 4
Gas concentration (ppm) and dynamic sensors response curve of ethylene-CO gases based on time (in seconds).
Fig. 5
Fig. 5
16 chemical gas sensor array response for the mixture of ethylene-methane.
Fig. 6
Fig. 6
Gas concentration (ppm) and dynamic sensors response curve of ethylene-methane gases based on time (in seconds).
Fig. 7
Fig. 7
Data division model in the training and testing set.
Fig. 8
Fig. 8
PCA of dataset features.
Fig. 9
Fig. 9
Performance evaluation of accuracy (%) and precision (%) based on tree-based ML approach on dynamic gas sensor dataset.
Fig. 10
Fig. 10
Performance evaluation of recall (%) and specificity (%) based on tree-based ML approach on dynamic gas sensor dataset.
Fig. 11
Fig. 11
Permutation importance analysis.
Fig. 12
Fig. 12
Confusion matrices of the model.
Fig. 13
Fig. 13
Performance evaluation of F1-score (%) and comparative analysis based on tree-based ML approach on dynamic gas sensor dataset.
Fig. 14
Fig. 14
Comparative performance analysis based on the existing tree-based ML models.
See this image and copyright information in PMC

References

    1. Gardner, J. W., Shin, H. W. & Hines, E. L. An electronic nose system to diagnose illness. Sens. Actuators B Chem.70, 1–3 (2000).
    1. Liu, T. et al. Decay characteristic of gas Arc in C₄F₇N/N₂ and C₄F₇N/CO₂ gas mixture by Thomson scattering. J. Phys. D. 58 (4), 045203 (2025).
    1. Meng, X. et al. Identification of thermal fault States in cable insulation sheaths based on gas sensor arrays. IEEE Trans. Dielectr. Electr. Insul.10.1109/TDEI.2025.3572332 (2025).
    1. Stetter, J. R., Jurs, P. C. & Rose, S. L. Detection of hazardous gases and vapors: pattern recognition analysis of data from an electrochemical sensor array, Anal. Chem., 58 (4), 860–866. 10.1021/ac00295a047 (1986).
    1. Maier, D. E., Hulasare, R., Qian, B. & Armstrong, P. Monitoring carbon dioxide levels for early detection of spoilage and pests in stored grain, in Proceedings of the 9th International Working Conference on Stored Product Protection, 117 (2006).

LinkOut - more resources