Prediction of hydraulic conductivity based on the soil grain size using supervised committee machine artificial intelligence

Mohammad Khalili-Maleki,Ramin Vafaei Poursorkhabi,Ata Allah Nadiri,Rouzbeh Dabiri
DOI: https://doi.org/10.1007/s12145-022-00848-x
2022-09-06
Earth Science Informatics
Abstract:Hydraulic conductivity is one of the important parameters to simulate water flow in porous environments that is suffer from uncertainty due to the different estimating approaches such as field and laboratory tests. Recently, artificial intelligence models have been applied to estimate hydrogeological parameters based on the available data. In this study, three individual artificial intelligence models including Larsen Fuzzy Logic (LFL), Least Square Support Vector Machine (LSSVM) and Hybrid Wavelet-Artificial Neural Network (WANN) were adopted to estimate hydraulic conductivity in Tabriz Urban Train based on grain size data. For optimal application of the advantages of the individual models, their output was used as the input of a nonlinear combiner. Comparison of the prediction results of this multiple model called Supervised Committee Machine Artificial Intelligence (SCMAI) with the individual models showed a 26% increase in Determination Coefficient (R 2 ) and a 36% decrease in Root Mean Squared Error (RMSE) compared to the most accurate individual artificial intelligence model, i.e., WANN. The results showed that the SCMAI model has a better performance in predicting hydraulic conductivity compared to the individual models and the superiority of the SCMAI model is due to using the type of individual models and the use of special advantages and strengths of each of these models.
geosciences, multidisciplinary,computer science, interdisciplinary applications
What problem does this paper attempt to address?