Genetic Algorithm-Optimized Deep Recurrent Networks for PM۲.۵ Forecasting: A Case Study on Urban Air Pollution in Tehran

Air pollution, especially fine particulate matter (PM۲.۵), remains a major public health and environmental concern in megacities like Tehran. This study proposes a hybrid deep learning framework that integrates recurrent neural networks (RNN, LSTM, GRU) with Genetic Algorithm-based hyperparameter optimization to forecast PM۲.۵ concentrations. To address missing values and inconsistent timestamps in the raw monitoring data, we reconstructed a complete and regular hourly-resolution time series by generating all ۲۴ hourly records for each station and applying forward- and backward-filling imputation. Additionally, we introduced station-wise embeddings to capture spatial variability across monitoring sites. The dataset was divided into ۸۰% training and ۲۰% testing, with ۲۰% of the training data further reserved for validation. The proposed method significantly outperformed baseline models, achieving a Mean Absolute Error (MAE) of ۰.۰۰۴۵, Root Mean Square Error (RMSE) of ۰.۰۱۰۲, and R² score of ۰.۹۹۶۶. These findings emphasize the effectiveness of evolutionary hyperparameter tuning and spatio-temporal modeling for air quality prediction systems.

Keywords:

air quality prediction , pm۲.۵ forecasting , recurrent neural networks , genetic algorithm , hyperparameter optimization