Search

In recent times, significant research has been carried out into developing and applying soft computing techniques for modeling hydro-climatic processes such as seepage modeling. It is necessary to properly model seepage, which creates groundwater sources, to ensure adequate management of scarce water resources. On the other hand, excessive seepage can threaten the stability of earthfill dams and infrastructures. Furthermore, it could result in severe soil erosion and consequently cause environmental damage. Considering the complex and nonlinear nature of the seepage process, employing soft computing techniques, especially applying pre-post processing techniques as hybrid methods, such as wavelet analysis, could be appropriate to enhance modeling efficiency. This review paper summarizes standard soft computing techniques and reviews their seepage modeling and simulation applications in the last two decades. Accordingly, 48 research papers from 2002 to 2021 were reviewed. According to the reviewed papers, it could be understood that regardless of some limitations, soft computing techniques could simulate the seepage successfully either through groundwater or earthfill dam and hydraulic structures. Moreover, some suggestions for future research are presented. This review was conducted employing preferred reporting items for systematic reviews and meta-analyses (PRISMA) method. [ABSTRACT FROM AUTHOR]

The aim of ensemble precipitation prediction in this paper was to achieve the best performance via artificial intelligence (AI) based modeling. In this way, ensemble AI based modeling was proposed for prediction of monthly precipitation with three different AI models (feed forward neural network-FFNN, adaptive neural fuzzy inference system-ANFIS and least square support vector machine-LSSVM) for the seven stations located in the Turkish Republic of Northern Cyprus (TRNC). Two scenarios were examined each having specific inputs set. The scenario 1 was developed for predicting each station's precipitation through its own data at previous time steps while in scenario 2, the central station's data were imposed into the models, in addition to each station's data, as exogenous input. Afterwards, the ensemble modeling was generated to improve the performance of the precipitation predictions. To end this aim, two linear and one non-linear ensemble techniques were used and then the obtained outcomes were compared. In terms of efficiency measures, the averaging methods employing scenario 2 and non-linear ensemble method revealed higher prediction efficiency. Also, in terms of Skill score, non-linear neural ensemble method could enhance predicting efficiency up to 44% in the verification step. [ABSTRACT FROM AUTHOR]