Publication: Application of Artificial Intelligence Models for modeling Water Quality in Groundwater: Comprehensive Review, Evaluation and Future Trends
Date
2021
Authors
Hanoon M.S.
Ahmed A.N.
Fai C.M.
Birima A.H.
Razzaq A.
Sherif M.
Sefelnasr A.
El-Shafie A.
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Science and Business Media Deutschland GmbH
Abstract
This study reported the state of the art of different artificial intelligence (AI) methods for groundwater quality (GWQ) modeling and introduce a brief description of common AI approaches. In addtion a bibliographic review of practices over the past two decades, was presented and attained result were compared. More than 80 journal articles from 2001 to 2021 were review in terms of characteristics and capabilities of developing methods, considering data of input-output, etc. From the reviewed studies, it could be concluded that in spite of various weaknesses, if the artificial intelligence approaches were appropriately built, they can effectively be utilized for predicting the GWQ in various aquifers. Because many steps of applying AI methods are based on trial-and-error or experience procedures, it�s helpful to review them regarding the special application for GWQ modeling. Several partial and general findings were attained from the reviewed studies that could deliver relevant guidelines for scholars who intend to carry out related work. Many new ideas in the associated area of research are also introduced in this work to develop innovative approaches and to improve the quality of prediction water quality in groundwater for example, it has been found that the combined AI models with metaheuristic optimization are more reliable in capturing the nonlinearity of water quality parameters. However, in this review few papers were found that used these hybrid models in GWQ modeling. Therefore, for future works, it is recommended to use hybrid models to more furthere investigation and enhance the reliability and accuracy of predicting in GWQ. � 2021, The Author(s), under exclusive licence to Springer Nature Switzerland AG.
Description
Aquifers; Forecasting; Groundwater resources; Hydrogeology; Machine learning; ANN; Artificial intelligence; Artificial intelligence methods; Future trends; Groundwater quality; Hybrid model; Intelligence models; Machine learning; Quality modeling; Water quality; boron; chloride; fluoride; ground water; nitrate; phosphate; sulfate; zinc; accuracy assessment; aquifer; artificial intelligence; future prospect; groundwater; guideline; machine learning; prediction; reliability analysis; trend analysis; water quality; accuracy; alkalinity; artificial intelligence; artificial neural network; chemical oxygen demand; concentration (parameter); electric conductivity; Escherichia coli; feed forward neural network; fuzzy logic; human; machine learning; multilayer perceptron; nonhuman; pH; physical parameters; practice guideline; prediction; radial basis function; radial basis function neural network; Review; single layer perceptron; support vector machine; suspended particulate matter; total dissolved solid; total hardness; trend study; turbidity; water quality