A novel soft sensor based warning system for hazardous ground-level ozone using advanced damped least squares neural...

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Data Science

Background:

• Hazardous air pollutants, particularly ground-level ozone, pose significant risks to public safety in urban environments.
• Accurate estimation and early warning systems for ozone are crucial for mitigating health impacts.

Purpose of the Study:

• To develop an efficient and low-cost air quality warning system for hazardous ground-level ozone.
• To estimate ground-level ozone concentrations using meteorological factors and a novel soft sensor approach.

Main Methods:

• Developed a ground-level ozone soft sensor using a damped least squares neural network (DLSNN) with greedy backward elimination (GBE).
• Utilized three key meteorological factors as input variables for ozone estimation.
• Implemented a weighted k-nearest neighbors (WkNN) classifier for the air quality warning system.
• Analyzed seasonal variations and statistical correlations of ozone concentration in Taiwan's urban areas.

Main Results:

• The DLSNN/GBE method demonstrated superior performance with low Mean Square Error (MSE) and Mean Absolute Error (MAE), and a high coefficient of determination (R²).
• A strong fit was achieved in determining ozone concentration from atmospheric meteorological features.
• The WkNN classifier achieved a high F1-score of 0.952, indicating excellent performance for the air quality warning system.

Conclusions:

• The proposed soft sensor and warning system provide an efficient and accurate method for estimating hazardous ground-level ozone.
• The system effectively utilizes meteorological data for real-time air quality monitoring and public safety.
• This approach offers a robust and cost-effective solution for urban air quality management.