Fine-tuning inflow prediction models: integrating optimization algorithms and TRMM data for enhanced accuracy

• Published in: Water science and technology 2024-08, Vol.90 (3), p.844-877
• Main Authors: Ali, Enas, Zerouali, Bilel, Tariq, Aqil, Katipoğlu, Okan Mert, Bailek, Nadjem, Santos, Celso Augusto Guimarães, M Ghoneim, Sherif S, Towfiqul Islam, Abu Reza Md
• Format: Article
• Language: English
• Subjects: Algorithms; data-driven frameworks; discrete wavelet transform; inflow prediction; Machine Learning; Models, Theoretical; parameter optimization; particle swarm optimization; reservoir management
• ISSN: 0273-1223; 1996-9732
• DOI: 10.2166/wst.2024.222

This research explores machine learning algorithms for reservoir inflow prediction, including long short-term memory (LSTM), random forest (RF), and metaheuristic-optimized models. The impact of feature engineering techniques such as discrete wavelet transform (DWT) and XGBoost feature selection is investigated. LSTM shows promise, with LSTM-XGBoost exhibiting strong generalization from 179.81 m /s RMSE (root mean square error) in training to 49.42 m /s in testing. The RF-XGBoost and models incorporating DWT, like LSTM-DWT and RF-DWT, also perform well, underscoring the significance of feature engineering. Comparisons illustrate enhancements with DWT: LSTM and RF reduce training and testing RMSE substantially when using DWT. Metaheuristic models like MLP-ABC and LSSVR-PSO benefit from DWT as well, with the LSSVR-PSO-DWT model demonstrating excellent predictive accuracy, showing 133.97 m /s RMSE in training and 47.08 m /s RMSE in testing. This model synergistically combines LSSVR, PSO, and DWT, emerging as the top performers by effectively capturing intricate reservoir inflow patterns.