Publication

Abstract : Over the last decades, air pollution has become one of the most serious environmental concerns. In 2021, Kolkata ranked as India's second most polluted city and 60th in the world. The ability to predict air quality is essential for warning the public about its levels. This will enable us to develop the ideal system for reducing the adverse effects of poor air quality on people's health. An Interval Type-3 Fuzzy Logic predictive system has been created to predict the AQI of Kolkata city of India. To improve the accuracy, the EKF learning algorithm has been taken. By this model, pollutant uncertainty and nonlinearity behavior will be mitigated. The root mean square error (RMSE), correlation coefficient (R2), mean absolute error (MAE) and mean square relative error (MSRE) have been employed to test the accuracy of prediction models. Data from the Central Pollution Control Board of India have been used to predict AQI. The suggested approach outperforms previous prediction models, offering more accuracy and prediction rates (RMSE: 0.044, R:0.977389). This work also demonstrates how effectively and conveniently merging advanced fuzzy logic with air quality forecasting may be used to address a variety of relevant environmental issues.