We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Modeling the total hardness (TH) of groundwater in aquifers using novel hybrid soft computing optimizer models.
- Authors
Moayedi, Hossein; Salari, Marjan; Ali, Sana Abdul-Jabbar; Dehrashid, Atefeh Ahmadi; Azadi, Hossein
- Abstract
A groundwater reservoir is either a solitary aquifer or a network of interconnected aquifers. A particular aquifer's groundwater purity evaluation could be time-consuming and costly. This study quantified the properties of Na%, SO42−, Cl, Na+, Mg2+, Ca2+, HCO3−, K+, and pH to predict the water quality parameter known as total hardness (as CaCO3). Groundwater quality data for the Shiraz Plain from 2002 to 2018 was utilized to accomplish this objective. The paper contrasts a hybrid methodology that combines Teaching Learning-Based Optimization (TLBO), Multiverse Optimizer (MVO), the Cuckoo Optimization Algorithm (COA), and the Evaporation Rate-based Water Cycle Algorithm (ER-WCA) with Artificial Neural Networks (ANN) this was done to design an optimal network for groundwater quality with conventional ANN. In comparison to all other TLBO-ANN, MVO-ANN, and COA-ANN developed models, the ER-WCA-ANN technique (with a population size of 500 and eight neurons in each hidden layer) provided the most accurate prediction for the TH with R2 values of 0.9983 and 0.98261, and RMSE values of 0.03698 and 0.00611, respectively, in the training and testing datasets. A comparison of the findings for the forecasting of groundwater quality showed that the ER-WCA-ANN hybrid model might increase prediction accuracy. These findings may have significant implications for future groundwater quality assessments.
- Subjects
SOFT computing; GROUNDWATER quality; OPTIMIZATION algorithms; GROUNDWATER; ARTIFICIAL neural networks; GROUNDWATER monitoring
- Publication
Environmental Earth Sciences, 2024, Vol 83, Issue 13, p1
- ISSN
1866-6280
- Publication type
Article
- DOI
10.1007/s12665-024-11618-x