AI application of the outlier-robust extreme learning machine (ORELM) model in river discharge prediction

Document Type : Original Article

Authors

1 Department of Computer Engineering, Razi University, Kermanshah, Iran.

2 Department of Water Engineering, Razi University, Kermanshah, Iran.

3 Department of Civil Engineering, Razi University, Kermanshah, Iran.

Abstract

Objective: This study focuses on evaluating the effectiveness of the Outlier-robust extreme learning machine (ORELM) model in predicting the discharge of the Qarasu River at the Pol Kohneh hydrological station in Kermanshah.
 
Method: The study focuses on the Pol Kohneh watershed along the Qarasu River, which originates from Sarab Ravansar, 50 kilometers northwest of Kermanshah. This study applies ELM to predict rainfall and river discharge using hydrological and meteorological data. ELM’s ability to handle complex, nonlinear relationships makes it suitable for forecasting, especially in real-time applications. To improve accuracy, outlier removal is performed before training. The model efficiently predicts future discharge values, offering a scalable solution for water resource management and flood risk assessment.
 
Results: The RMSE values for a one-month lag (t-1) were recorded as 12.22 and 4.14 for the training and testing phases, respectively. The findings of this study highlight the potential of artificial intelligence-based models like ORELM in hydrological forecasting. Unlike traditional methods, these models require significantly less time and computational effort while delivering accurate predictions. By leveraging machine learning techniques, water resource managers can efficiently forecast river discharge trends, aiding in effective decision-making for flood management, irrigation planning, and water allocation during dry and wet periods.
 
Conclusions:  The ability of ORELM to handle long-term discharge predictions with minimal input data makes it a valuable tool for hydrological studies, particularly in data-scarce regions where conventional models face limitations.

Keywords

Main Subjects

References

Bărbulescu, A., & Zhen, L. (2024). Forecasting the river water discharge by artificial intelligence methods. Water, 16(9), 1248. https://doi.org/10.3390/w16091248
Darabi Cheghabaleki, S., Fatemi, S. E., & Hafezparast Mavadat, M. (2024). Enhancing spatial streamflow prediction through machine learning algorithms and advanced strategies. Applied Water Science, 14(6), 110. http://dx.doi.org/10.1007/s13201-024-02154-x
Defontaine, T., Ricci, S., Lapeyre, C. J., Marchandise, A., & Le Pape, E. (2024). Real-time flood forecasting with Machine Learning using scarce rainfall-runoff data. EGUsphere, 1-32. https://doi.org/10.5194/egusphere-2023-2621
Deo, R. C., Downs, N., Parisi, A. V., Adamowski, J. F., & Quilty, J. M. (2017). Very short-term reactive forecasting of the solar ultraviolet index using an extreme learning machine integrated with the solar zenith angle. Environmental research, 155, 141-166. https://doi.org/10.1016/j.envres.2017.01.035
Fatemi, S. E., & Parvini, H. (2022). The impact assessments of the ACF shape on time series forecasting by the ANFIS model. Neural Computing and Applications, 34(15), 12723-12736. http://dx.doi.org/10.1007/s00521-022-07140-5
Gorgani, S., Bafkar, A., & Fatemi, S. (2017). Prediction of groundwater pollution potential using the DRASTIC index and annual time series analysis (case study: plain Mahidasht Kermanshah). IJHE, 10 (3), 317-328. http://ijhe.tums.ac.ir/article-1-5962-en.html (In Persian)
Huang, G., Huang, G.-B., Song, S., & You, K. (2015). Trends in extreme learning machines: A review. Neural Networks, 61, 32-48. http://dx.doi.org/10.1016/j.neunet.2014.10.001
Huang, G.-B., Zhu, Q.-Y., & Siew, C.-K. (2006). Extreme learning machine: theory and applications. Neurocomputing, 70(1-3), 489-501. https://doi.org/10.1016/j.neucom.2005.12.126
Morovati, R., & Kisi, O. (2024). Utilizing hybrid machine learning techniques and gridded precipitation data for advanced discharge simulation in under-monitored river basins. Hydrology, 11(4), 48. https://doi.org/10.3390/hydrology11040048
Raja Shekar, P., & Mathew, A. (2024). AI-Based Rainfall-Runoff Modelling for Sustainable Water Management in Potteruvagu Watershed, India. In Geospatial Technology to Support Communities and Policy: Pathways to Resiliency, Springer, Springer, New York Sity, United states. http://dx.doi.org/10.1007/978-3-031-52561-2_6
Salcedo-Sanz, S., Pastor-Sánchez, A., Prieto, L., Blanco-Aguilera, A., & García-Herrera, R. (2014). Feature selection in wind speed prediction systems based on a hybrid coral reefs optimization–Extreme learning machine approach. Energy Conversion and Management, 87, 10-18. https://doi.org/10.1016/j.enconman.2014.06.041
Wang, E., Zhang, Y., Luo, J., Chiew, F. H. S., & Wang, Q. J. (2011). Monthly and seasonal streamflow forecasts using rainfall-runoff modeling and historical weather data. Water Resources Research, 47(5). https://doi.org/https://doi.org/10.1029/2010WR009922
Zhang, X., Wang, R., Wang, W., Zheng, Q., Ma, R., Tang, R., & Wang, Y. (2025). Runoff prediction using combined machine learning models and signal decomposition. Journal of Water and Climate Change, 16(1), 230-247. https://doi.org/10.2166/wcc.2024.663
 
Advanced Technologies in Water Efficiency
Volume 5, Issue 3
September 2025
Pages 44-55

Files

History

  • Receive Date: 11 December 2024
  • Revise Date: 03 February 2025
  • Accept Date: 15 March 2025

Share

How to cite

Statistics