Air quality prediction and control systems using machine learning and adaptive neuro-fuzzy inference system

Area of Science:

• Environmental engineering
• Artificial intelligence applications in environmental science
• Air quality monitoring and prediction

Background:

• Air quality prediction is complex due to nonlinear pollutant interactions.
• Accurate forecasting is crucial for environmental management and public health.
• Existing methods may not fully capture the dynamic nature of air pollution.

Purpose of the Study:

• To develop and evaluate artificial intelligence models for air quality forecasting in Semnan, Iran.
• To identify the most reliable machine learning model for predicting pollutant concentrations.
• To establish a robust framework for environmental engineering air quality assessments.

Main Methods:

• Collected and analyzed comprehensive data for seven different air pollutants.
• Evaluated the performance of several machine learning (ML) models.
• Focused on the adaptive neuro-fuzzy inference system (ANFIS) for its predictive capabilities.

Main Results:

• The adaptive neuro-fuzzy inference system (ANFIS) demonstrated superior performance in air quality prediction.
• ANFIS showed high reliability across various datasets and seasons.
• Machine learning models were rigorously assessed for forecasting accuracy.

Conclusions:

• ANFIS is a highly effective tool for seasonal air quality prediction in the studied region.
• Artificial intelligence offers a reliable framework for advanced environmental engineering solutions.
• The findings provide a valuable case study for air quality management in Iran.