Authors
1
Doctoral student in water resources management civil engineering, Tabriz University
2
Faculty of Civil Engineering, University of Tabriz
3
Assistant Prof., Dept. of Water Resources Eng., Faculty of Civil Eng.,Univ. of Tabriz,Iran.
10.22034/ceej.2024.59627.2308
Abstract
Based on the vital role of water in human life, precise appraisal of the potential of water resources and their optimum use is a significant issue in scientific circles and the water industry. In this paper, the ability of the artificial neural network model with error back propagation and other models based on data mining in the Weka software platform is used in forecasting modeling a few steps later using observational inputs. At the end, the performance of all models has been checked using evaluation criteria and the models have been compared with each other. The purpose of this paper is to forecast the incoming flow to the Alavian reservoir in the next few months. In this research, the model inputs include the runoff entering the dam reservoir, temperature, evaporation, precipitation and snow cover of the dam catchment area and drought indicators, and these data are monthly and It has been used for 25 years (1997-2022). Comparing the results of the ANN model with other RF, RT, GP, and SM models in the verification stage, it shows that in all steps, the prediction results were suitable for all models, and among the 5 models whose results were compared, the average results of the coefficient The correlation of ANN, RF, and RT models is 85%, 90%, and 83%, respectively, compared to other models in predicting the inflow to the Alavian reservoir in the next few steps , suitable and close to the observational data, especially at the peak points have been.
Keywords
Main Subjects
)