One Step Ahead Prediction of Ozone Concentration for Determination of Outdoor Air Quality Level

Authors : Waleed MAHMOOD, Ercan AVŞAR

Pages : 45-54

Doi:10.51354/mjen.869736

View : 19 | Download : 9

Publication Date : 2021-06-30

Article Type : Research Paper

Abstract :With the rapid spread of urbanization, competent authorities become increasingly anxious from air pollution risks and effect on citizens especially those with respiratory diseases. In this work, performances of six machine learning methods were analyzed for prediction of maximum ozone insert ignore into journalissuearticles values(O_3); concentration for the next-day. The models make the prediction using concentrations of six atmospheric components insert ignore into journalissuearticles values(PM2.5, PM10, Ozone insert ignore into journalissuearticles values(O3);, Sulfur Dioxide insert ignore into journalissuearticles values(SO2);, Nitrogen Dioxide insert ignore into journalissuearticles values(NO2);, and Carbon Monoxide insert ignore into journalissuearticles values(CO););. The utilized machine learning methods are multilayer perception insert ignore into journalissuearticles values(MLP);, Support Vector Regression insert ignore into journalissuearticles values(SVM);, k-Nearest Neighbor insert ignore into journalissuearticles values(K-NN);, Random Forests insert ignore into journalissuearticles values(RF);, Gradient Boosting insert ignore into journalissuearticles values(GB);, and Elastic Net insert ignore into journalissuearticles values(EN);. After the predictions made by these models, the predicted values were further processed to be classified into one of the six air quality levels defined by United States Environmental Protection Agency. The prediction performances of the models as well as their corresponding classification results were analyzed. It was shown that MLP model gives the lowest RMSE of 2246 for prediction step while SVR achieved the highest accuracy score of 0.790.
Keywords : machine learning, time series forecasting, regression methods, sequence to sequence, air quality index