Breakthrough in Atmospheric Pollution Detection: Researchers Develop AI-Powered System

Strategic Insights -

Researchers at Yanshan University in China have made a significant contribution to the field of atmospheric pollution detection with the development of an AI-powered online detection system. The system, based on ultraviolet differential optical absorption spectroscopy combined with convolutional attention neural network, is capable of detecting multiple components of atmospheric pollution, including sulfur dioxide, carbon disulfide, and nitrogen oxides, with high precision. The study has been funded by the National Natural Science Foundation of China and other organizations.

Key Takeaways:

  • The system uses a combination of UV-DOAS and CANN to detect multiple components of atmospheric pollution with high precision.
  • The detection system can achieve an R2 value of 0.9995 for SO2, 0.9995 for CS2, and 0.9993 for NO in the range of 0.1-10.02 ppm.
  • The RMSE values for SO2, CS2, and NO were reduced to 0.070 ppm, 0.077 ppm, and 0.081 ppm, respectively.
  • The system has a high sensitivity, with detection limits of 4.7 ppb center dot m, 6.3 ppb center dot m, and 5.6 ppb center dot m for SO2, CS2, and NO, respectively.
  • The research has been peer-reviewed and published in the journal Sensors and Actuators B: Chemical.

Statistics:

  • R2 values for SO2, CS2, and NO: 0.9995, 0.9995, and 0.9993, respectively.
  • RMSE values for SO2, CS2, and NO: 0.070 ppm, 0.077 ppm, and 0.081 ppm, respectively.
  • Detection limits for SO2, CS2, and NO: 4.7 ppb center dot m, 6.3 ppb center dot m, and 5.6 ppb center dot m, respectively.

Sources:

  • NewsRx. New Global Warming and Climate Change Study Findings Have Been Reported by Investigators at Yanshan University (An Online Trace Gas Detection System for Multi-component Atmospheric Pollutants Based On Ultraviolet Differential Optical Absorption ...). Global Warming Focus. October 20, 2025; p 1999.
  • An Online Trace Gas Detection System for Multi-component Atmospheric Pollutants Based On Ultraviolet Differential Optical Absorption Spectroscopy Combined With Convolutional Attention Neural Network. Sensors and Actuators B: Chemical, 2025; 441.