Document Type : Research Article
Authors
1
Department of GIS/RS, Faculty of Natural Resources and Environment, Science and Research Branch, Islamic Azad University, Tehran, Iran
2
Department of GIS/RS, Faculty of Environment and Energy, Science and Research Branch, Islamic Azad University,Tehran,Iran
3
Department of Civil Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran
10.22059/ije.2024.374112.1806
Abstract
Limited groundwater resources and their overuse have become fundamental challenges for sustainable development worldwide. In this study, a combination of step-by-step weighted evaluation ratio analysis (SWARA), deep learning method of convolutional neural network (CNN), whale optimization (WOA) and bee colony algorithms (ABC) be applied, which approach provides an innovative method to produce the groundwater spring potential in Kermanshah province. In the first stage, a map containing 301 spring locations with a flow rate of more than 30 cubic meters per second and 304 points without springs was prepared. Thirteen parameters were created as independent variables. The SWARA method was used to determine the weight of the criteria, decision-making indicators and determine the relationship between the spring creation potential and the selected factors. Then, CNN-WOA hybrid model be applied to prepare the groundwater spring potential, and then it will be evaluated from the performance characteristic curve (ROC) and some other statistical evaluations. The validation of the training dataset illustrated that the success rate for SWARA-CNN-WOA, SWARA-CNN-ABC models is 86%, 91%, respectively. The results showed that the SWARA-CNN-ABC model performed better than other models with a small difference. In addition, the prediction rate evaluation revealed that the values under the ROC curve for SWARA-CNN-ABC, SWARA-CNN-WOA models are 87%, 88%, respectively. Based on the results, despite the excellent performance of all models, the SWARA-CNN-ABC model has made more accurate predictions. The hybrid models presented in this study can be used as an efficient and effective methodology to improve groundwater potential.
Keywords
Main Subjects
)