EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Research on Binary Mixed VOCs Gas Identification Method Based on Multi-Task Learning

Haixia Mei iD, Ruiming Yang, Jingyi Peng, Keyu Meng, Tao Wang iD, Lijie Wang

Sensors · 08.04.2025 · 25 (8) · 2355

DOI: 10.3390/s25082355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 08.04.2025. Verfasst von Haixia Mei und 5 weiteren Autor:innen. Beteiligte Einrichtung: Key Lab Intelligent Rehabil & Barrier Free Disable (Ministry of Education), Changchun University, Changchun 130022, China. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jilin Provincial Natural Science Foundation, National Natural Science Foundation of China, Jilin Provincial Science and Technology Development Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional volatile organic compounds (VOCs) detection models separate component identification and concentration prediction, leading to low feature utilization and limited learning in small-sample scenarios. Here, we realize a Residual Fusion Network based on multi-task learning (MTL-RCANet) to implement component identification and concentration prediction of VOCs. The model integrates channel attention mechanisms and cross-fusion modules to enhance feature extraction capabilities and task synergy. To further balance the tasks, a dynamic weighted loss function is incorporated to adjust weights dynamically according to the training progress of each task, thereby enhancing the overall performance of the model. The proposed network achieves an accuracy of 94.86% and an R2 score of 0.95. Comparative experiments reveal that using only 35% of the total data length as input data yields excellent identification performance. Moreover, multi-task learning effectively integrates feature information across tasks, significantly improving model efficiency compared to single-task learning.

Schlagwörter

Stichwörter der Autor:innen

feature fusiongas sensormixed gasesmulti-task learning

Autor:innen und Institutionen

  • Key Lab Intelligent Rehabil & Barrier Free Disable (Ministry of Education), Changchun University, Changchun 130022, China.
  • State Key Laboratory of Integrated Optoelectronics, College of Electronic Science and Engineering, Jilin University, Changchun 130012, China.
  • Shanghai Key Laboratory of Intelligent Sensing and Detection Technology, School of Mechanical and Power Engineering, East China University of Science and Technology, Shanghai 200237, China.

ORCID-Profile: Haixia Mei, Tao Wang

Förderung

  • Jilin Provincial Natural Science Foundation
  • National Natural Science Foundation of China
  • Jilin Provincial Science and Technology Development Program

Zitieren

APA

Mei, H., Yang, R., Peng, J., Meng, K., Wang, T., & Wang, L. (2025). Research on Binary Mixed VOCs Gas Identification Method Based on Multi-Task Learning. Sensors, 25(8), 2355. https://doi.org/10.3390/s25082355

Vancouver

Mei H, Yang R, Peng J, Meng K, Wang T, Wang L. Research on Binary Mixed VOCs Gas Identification Method Based on Multi-Task Learning. Sensors. 2025;25(8):2355. doi:10.3390/s25082355

BibTeX
@article{Mei2025,
  author = {Mei, Haixia and Yang, Ruiming and Peng, Jingyi and Meng, Keyu and Wang, Tao and Wang, Lijie},
  title = {Research on Binary Mixed VOCs Gas Identification Method Based on Multi-Task Learning},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {8},
  pages = {2355},
  doi = {10.3390/s25082355}
}
Kostenlos starten →